Bridge flooding sensing and predicting system and method

Abstract

The invention provides a bridge flooding sensing prediction system and method, which is applied to the technical field of bridge flooding sensing, and comprises a data acquisition module used for acquiring multi-source monitoring data of a bridge in real time; the data processing module is used for generating bridge state fusion parameters through a multi-source sensing dynamic fusion model based on the water level data, the geological displacement data and the bridge floor water regimen image data; on the basis of the bridge state fusion parameters, combining the main stream hydrological data with the branch hydrological data, and predicting flooding risk parameters of a downstream bridge through a hydrological dynamic evolution model; the data transmission module is used for transmitting the flooding risk parameters to a pre-associated early warning execution module through dual-channel redundancy; and the early warning execution module is used for executing local sound-light alarm through a warning lamp and a loudspeaker when the flooding risk parameter exceeds a risk threshold value, and pushing remote early warning information to a preset user through a short message platform. According to the invention, the accuracy and real-time performance of bridge flooding early warning can be improved.

Description

Technical Field

[0001] This invention relates to the field of bridge flood sensing technology, and in particular to a bridge flood sensing and prediction system and method. Background Technology

[0002] Current bridge flooding detection and prediction technologies mainly rely on single sensor monitoring, traditional communication methods, or retrospective manual inspections, which suffer from problems such as slow response, insufficient accuracy, and limited coverage.

[0003] For example, bridge flood monitoring methods based on a single water level sensor involve deploying contact-type water level gauges (such as float-type or pressure-type) at key locations on the bridge and transmitting data to a local platform via wired networks or cellular communication. However, a single water level sensor can only acquire point-specific data and cannot detect flood risks across the entire area (such as water accumulation in low-lying areas of the bridge deck or slope erosion), and it also has a relatively large error margin.

[0004] It is evident that the bridge flood sensing methods in related technologies have significant limitations and insufficient accuracy. Summary of the Invention

[0005] This invention provides a bridge flooding sensing and prediction system and method to address the shortcomings of existing bridge flooding sensing methods, such as limited coverage and insufficient accuracy, thereby improving the accuracy and real-time performance of bridge flooding early warning.

[0006] This invention provides a bridge flooding perception and prediction system, comprising the following modules: a data acquisition module, used to acquire multi-source monitoring data of the bridge in real time, wherein the multi-source monitoring data includes: water level data collected by a water level sensor, geological displacement data of the bridge monitored by a GNSS receiver, and bridge deck water condition image data captured by a camera; a data processing module, communicatively connected to the data acquisition module, used to generate bridge state fusion parameters based on the water level data, the geological displacement data, and the bridge deck water condition image data through a multi-source perception dynamic fusion model; based on the bridge state fusion parameters, combined with main stream hydrological data and tributary hydrological data, to predict downstream bridge flooding risk parameters through a hydrological dynamic evolution model; a data transmission module, communicatively connected to the data processing module, used to transmit the flooding risk parameters to a pre-associated early warning execution module through dual-channel redundancy; the early warning execution module, communicatively connected to the data transmission module, used to trigger a graded response according to the flooding risk parameters: when the flooding risk parameters exceed a risk threshold, a local audible and visual alarm is executed through warning lights and a horn, and remote early warning information is pushed to preset users through an SMS platform.

[0007] According to the present invention, a bridge flooding perception and prediction system is provided, wherein the bridge deck water condition image data captured by the camera includes: a PTZ camera for dynamically monitoring changes in the surrounding environment of the bridge; and a bullet camera for identifying bridge deck water condition parameters through a visual algorithm, wherein the bridge deck water condition parameters include flooding depth and water flow velocity.

[0008] According to the present invention, a bridge flooding perception and prediction system is provided, wherein transmitting the flooding risk parameters to a pre-associated early warning execution module via dual-channel redundancy includes: real-time monitoring of wireless network signal strength; transmitting the flooding risk parameters to the pre-associated early warning execution module via BeiDou short message when the wireless network signal strength is less than a strength threshold; and transmitting the flooding risk parameters to the pre-associated early warning execution module via wireless network when the wireless network signal strength is greater than or equal to the strength threshold.

[0009] According to the bridge flooding detection and prediction system provided by the present invention, both the warning light and the horn are closed-loop circuit control devices, which are activated only when the flooding risk exceeds a threshold; both the warning light and the horn are driven by a battery powered by a photovoltaic panel.

[0010] According to the present invention, a bridge flooding perception and prediction system is provided. The system predicts downstream bridge flooding risk parameters based on the bridge state fusion parameters, combined with main stream hydrological data and tributary hydrological data, using a hydrological dynamic evolution model. The system includes: inputting the bridge state fusion parameters, the main stream hydrological data, and the tributary hydrological data into a trained hydrological dynamic evolution model for dynamic evolution modeling to obtain a predicted downstream flooding time; wherein the hydrological dynamic evolution model is used to correlate water level rise with flood propagation delay; and quantifying the risk of the predicted downstream flooding time to generate flooding risk parameters including the inundation range and evolution path.

[0011] According to the present invention, a bridge flooding perception and prediction system is provided. The step of generating bridge state fusion parameters based on the water level data, the geological displacement data, and the bridge deck water condition image data through a multi-source perception dynamic fusion model includes: dynamically associating the water level data and the geological displacement data to generate a geological displacement risk coefficient; and fusing the bridge positioning data, the geological displacement risk coefficient, and the bridge deck water condition image data to obtain the bridge state fusion parameters.

[0012] This invention also provides a method for predicting and sensing bridge flooding, comprising the following steps: acquiring multi-source monitoring data of the bridge in real time, wherein the multi-source monitoring data includes: water level data collected by a water level sensor, geological displacement data of the bridge monitored by a GNSS receiver, and bridge deck water condition image data captured by a camera; generating bridge state fusion parameters based on the water level data, the geological displacement data, and the bridge deck water condition image data through a multi-source sensing dynamic fusion model; predicting downstream bridge flooding risk parameters based on the bridge state fusion parameters, combined with main stream hydrological data and tributary hydrological data, through a hydrological dynamic evolution model; triggering a graded response according to the flooding risk parameters: when the flooding risk parameters exceed a risk threshold, executing a local audible and visual alarm through warning lights and a loudspeaker, and simultaneously pushing remote early warning information to preset users through an SMS platform.

[0013] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the bridge flooding sensing and prediction method as described above.

[0014] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the bridge flooding sensing and prediction method as described above.

[0015] The present invention also provides a computer program product, including a computer program that, when executed by a processor, implements the bridge flooding perception and prediction method as described above.

[0016] The bridge flooding sensing and prediction system and method provided by this invention collects water level, geological displacement, and bridge deck hydrological images in real time through multi-source monitoring data. A multi-source sensing dynamic fusion model generates bridge status fusion parameters, which, combined with hydrological data from the main stream and tributaries, utilize a hydrological dynamic evolution model to accurately predict the downstream bridge flooding risk. Dual-channel redundant transmission ensures data reliability, and finally, a tiered response is triggered based on the risk parameters, enabling local audible and visual alarms and remote information push notifications. The system achieves closed-loop intelligent management from monitoring and analysis to early warning, significantly improving the accuracy, real-time performance, and emergency response efficiency of bridge flooding early warning. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced one by one below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0018] Figure 1 This is a schematic diagram of the modules of the bridge flooding sensing and prediction system provided by the present invention.

[0019] Figure 2 This is a hardware structure diagram of the bridge flooding sensing and prediction system provided by the present invention.

[0020] Figure 3 This is a flowchart of the BeiDou-3 short message data transmission process provided by the present invention.

[0021] Figure 4 This is a flowchart illustrating the bridge flooding sensing and prediction method provided by the present invention.

[0022] Figure 5 This is a schematic diagram of the physical structure of the electronic device provided by the present invention. Detailed Implementation

[0023] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0024] Current bridge flooding detection and prediction technologies mainly rely on single sensor monitoring, traditional communication methods, or retrospective manual inspections, which have problems such as response lag, insufficient accuracy, and limited coverage. Below are some similar related cases.

[0025] In related technologies, monitoring systems based on a single water level sensor deploy contact-type water level gauges (such as float-type or pressure-type) at key locations on the bridge and transmit data to a local platform via wired networks or cellular communication.

[0026] However, a single water level sensor is susceptible to interference from rapid water flow and floating objects, and the water level measurement error usually exceeds ±5cm; it can only acquire point data and cannot detect the risk of flooding over the entire area (such as water accumulation in low-lying areas of the bridge deck and slope erosion).

[0027] In related technologies, BeiDou displacement monitoring involves deploying BeiDou receivers on bridges, monitoring water levels using millimeter-wave radar, and detecting deformation using structural diagnostic tools. Data is then uploaded via a local network and transmitted through a BeiDou communication module. Distributed hydrological station data is transmitted using BeiDou short message service, and communication efficiency is improved through resource allocation algorithms.

[0028] However, the multiple sensors are not integrated, and the water level and displacement data are analyzed independently, lacking a model for the causes of flooding (such as the displacement of bridge piers induced by rising water levels); the early warning is delayed, only responding to deformations that have already occurred, and failing to predict flooding risks; and the specificity of bridges is ignored: the sensing parameters designed for bridge structures (such as the depth of water in the bridge deck and the scouring rate of the piers) are not designed, resulting in a weak correlation between the data and flooding disasters.

[0029] Among the related technologies, the "space-air-ground" integrated monitoring system combines satellite remote sensing (large-scale screening), UAV inspection (high-risk point modeling) and ground sensors (BeiDou displacement gauges, inclinometers) to form a multi-level monitoring network.

[0030] However, it requires the deployment of a large amount of specialized equipment (such as deep inclinometers and millimeter-wave radar), with an investment of over one million yuan per bridge. Furthermore, the remote sensing data update cycle is as long as several hours, which cannot meet the minute-level flood warning requirements.

[0031] In related technologies, lightweight sensing integrates temperature, humidity, and vibration sensors, enabling them to be used immediately upon installation via solar power and wireless transmission.

[0032] However, lightweight sensing lacks water level sensing capabilities, only monitors structural posture, and cannot directly correlate with flooding events; moreover, it relies on historical data, requiring millions of reports to train the model, making it poorly adaptable to new bridges or sudden flood scenarios.

[0033] refer to Figure 1 , Figure 1 This is a schematic diagram of the bridge flooding perception and prediction system provided by the present invention, including: a data acquisition module 101, a data acquisition module 102, a data transmission module 103, and an early warning execution module 104.

[0034] The data acquisition module is used to acquire multi-source monitoring data of the bridge in real time. The multi-source monitoring data includes: water level data collected by a water level sensor, geological displacement data of the bridge monitored by a GNSS receiver, and bridge deck water condition image data captured by a camera. The data processing module is communicatively connected to the data acquisition module and is used to generate bridge state fusion parameters based on the water level data, the geological displacement data, and the bridge deck water condition image data through a multi-source sensing dynamic fusion model. Based on the bridge status fusion parameters, combined with the hydrological data of the main stream and the tributary, the flooding risk parameters of the downstream bridge are predicted through a hydrological dynamic evolution model. The data transmission module is communicatively connected to the data processing module and is used to transmit the flood risk parameters to the pre-associated early warning execution module through dual-channel redundancy. The early warning execution module is communicatively connected to the data transmission module and is used to trigger a graded response based on the flood risk parameter: when the flood risk parameter exceeds the risk threshold, a local audible and visual alarm is executed through warning lights and a horn, and at the same time, remote early warning information is pushed to preset users through the SMS platform.

[0035] In this embodiment of the invention, the data acquisition module is responsible for acquiring multi-source monitoring data of the bridge and its surrounding environment in real time.

[0036] Specifically, the data acquisition module continuously collects water level data at a preset sampling frequency using water level sensors deployed on bridge piers or in nearby waterways. Simultaneously, GNSS receivers fixedly installed on key bridge components (such as piers and abutments) continuously monitor millimeter-level geological displacement data of the bridge structure and its foundation. Furthermore, cameras mounted at both ends of the bridge or at higher elevations capture real-time video streams of the bridge deck and upstream waterway, extracting crucial bridge deck hydrological image data. This multi-source, heterogeneous data collectively forms the data foundation for the system's risk perception and prediction.

[0037] The data processing module and the data acquisition module have a stable communication connection to receive the aforementioned multi-source monitoring data.

[0038] The core processing of the data processing module is divided into two stages. In the first stage, the data processing module calls the built-in multi-source sensing dynamic fusion model to perform spatiotemporal alignment, feature extraction, and fusion calculation on water level data, geological displacement data, and bridge deck water condition image data, generating a bridge state fusion parameter that can comprehensively reflect the current safety status of the bridge.

[0039] In the second stage, the data processing module further integrates real-time hydrological data from external hydrological stations on the main stream and tributaries, inputting the bridge status fusion parameters along with the upstream and downstream hydrological conditions into the hydrological dynamic evolution model. This model simulates the evolution of the water flow and ultimately calculates the flooding risk parameters for the downstream target bridge. These parameters provide a quantitative assessment of the probability and severity of flooding in the future.

[0040] The data transmission module communicates with the data processing module. The core function of the data transmission module is to reliably transmit the calculated flood risk parameters to the early warning execution terminal. To ensure extremely high reliability of early warning information transmission, this module employs a dual-channel redundant transmission mechanism. That is, the same flood risk parameter data packet is simultaneously sent to the early warning execution module through at least two independent communication links (such as BeiDou short message service and 4G / 5G wireless network), ensuring that even if one link fails, the early warning information can still be delivered promptly and completely.

[0041] The early warning execution module communicates with the data transmission module. The early warning execution module is responsible for receiving flood risk parameters and executing the final early warning action. The early warning execution module has multiple preset risk thresholds, which can trigger corresponding tiered responses based on the received flood risk parameter values.

[0042] When the flood risk parameters exceed the highest risk threshold, the module will immediately trigger the highest level of emergency response: on the one hand, by driving the warning lights and loudspeakers set at the bridgehead or roadside, a strong local sound and light alarm will be executed to warn vehicles and pedestrians on site; on the other hand, simultaneously through the integrated high-speed SMS platform, remote early warning information containing risk level and location will be pushed in batches to the mobile phone numbers of users such as the pre-set maintenance unit and traffic management department leaders, thereby realizing a three-dimensional early warning system integrating on-site and remote monitoring.

[0043] This invention utilizes multi-source monitoring data to collect real-time images of water levels, geological displacements, and bridge deck water conditions. A multi-source sensing dynamic fusion model generates bridge status fusion parameters, which, combined with hydrological data from the main stream and tributaries, are used to accurately predict downstream bridge flooding risks using a hydrological dynamic evolution model. Dual-channel redundant transmission ensures data reliability, and a tiered response is triggered based on risk parameters, enabling local audible and visual alarms and remote information push notifications. The system achieves closed-loop intelligent management from monitoring and analysis to early warning, significantly improving the accuracy, real-time performance, and emergency response efficiency of bridge flooding warnings.

[0044] According to the present invention, a bridge flooding sensing and prediction system is provided, wherein the bridge deck water level image data captured by a camera includes: A PTZ camera is used to dynamically monitor changes in the surrounding environment of the bridge; A bullet camera is used to identify the bridge deck water condition parameters through visual algorithms, wherein the bridge deck water condition parameters include flood depth and water flow velocity.

[0045] In the data acquisition module, the specific implementation method for capturing bridge surface water condition image data by cameras includes deploying two complementary cameras.

[0046] The first type is a PTZ camera, which is typically mounted at a high point with a wide field of view, capable of covering the entire bridge and parts of the river upstream and downstream. This PTZ camera is equipped with a pan-tilt unit and automatic zoom, and is used to monitor changes in the surrounding environment of the bridge over a wide area and dynamically, such as water level rise trends, floating debris impacts, and the overall water flow of the river, providing the system with macroscopic visual situational awareness.

[0047] The second type is a bullet camera, which is typically fixedly mounted on the side of the bridge or on a light pole, with the lens aimed at a specific area of ​​the bridge deck. This bullet camera is used to identify the bridge deck's water level parameters through built-in or back-end deployed visual algorithms.

[0048] Specifically, the visual algorithm analyzes captured consecutive image frames, such as using a calibration reference method to calculate the height of the flooded area, thereby obtaining the flood depth; simultaneously, it analyzes the movement trajectory of floating objects on the water surface or calculates the displacement of water surface texture to obtain the water flow velocity. These quantitative bridge deck water condition parameters identified by the bullet camera provide crucial real-time observation data for subsequent data fusion and risk prediction.

[0049] In the data processing module, visual algorithms are used to analyze and process the bridge surface water condition images captured by the bullet camera. First, image preprocessing techniques, such as noise reduction and contrast enhancement, are used to improve image quality. Then, image segmentation algorithms are used to separate the waterlogged areas from the bridge surface background. Next, based on the relative positional relationship between known bridge surface reference points (such as a ruler set on the bridge surface or a fixed object of known height) and the waterlogged areas, combined with geometric transformations and proportional relationships in the image, the flood depth is calculated. For example, if the height of a ruler on the bridge surface is known to be H, the number of pixels occupied by the ruler in the image is P1, and the number of pixels from the water surface to a certain mark on the ruler is P2, the flood depth can be calculated by establishing the proportional relationship between pixels and actual length.

[0050] For water flow velocity identification, image preprocessing is performed first. Then, visual algorithms such as optical flow or particle image velocimetry are used. Optical flow estimates the velocity and direction of water flow by analyzing the motion vectors of pixels in the image. In bridge surface water condition images, water flow causes floating objects or water surface textures to move. Optical flow algorithms can track these motion features, calculate the velocity of each pixel, and thus obtain the velocity distribution of the entire water flow area. Particle image velocimetry involves releasing tracer particles (e.g., floating objects on the river surface or pre-set particles) into the water. By analyzing the displacement of the tracer particles in two consecutive frames of images and combining the time interval, the water flow velocity is calculated. By analyzing tracer particles at multiple locations, the average value or velocity field distribution of the water flow velocity can be obtained.

[0051] Through the embodiments of this invention, comprehensive monitoring of the surrounding environment and water conditions of the bridge deck is achieved by using PTZ cameras and bullet cameras in tandem. The dynamic monitoring function of the PTZ cameras can promptly detect potential hazards in the surrounding environment of the bridge, such as landslides and fallen trees that may affect bridge safety; while the bullet cameras focus on the accurate identification of water condition parameters on the bridge deck, providing detailed data support for assessing the safety of the bridge under flooded conditions.

[0052] According to the present invention, a bridge flooding sensing and prediction system is provided, wherein the flooding risk parameters are transmitted to a pre-associated early warning execution module via dual-channel redundancy, comprising: Real-time monitoring of wireless network signal strength; When the wireless network signal strength is less than the strength threshold, the flood risk parameters are transmitted to the pre-associated early warning execution module via BeiDou short message; When the wireless network signal strength is greater than or equal to the strength threshold, the flood risk parameter is transmitted to the pre-associated early warning execution module via the wireless network.

[0053] In this embodiment of the invention, the data transmission module is responsible for reliably transmitting flood risk parameters to the early warning execution module. The specific implementation of its dual-channel redundant transmission mechanism is as follows: First, the data transmission module has a built-in or connected signal monitoring unit for real-time monitoring and evaluation of the signal strength of conventional wireless networks (such as 4G / 5G cellular networks) to ensure continuous awareness of the primary communication link status.

[0054] The transmission strategy dynamically and intelligently switches based on the real-time monitored wireless network signal strength. Specifically, when the wireless network signal strength is lower than a preset strength threshold, it indicates that the primary wireless communication link is of poor quality or interrupted. At this time, the data transmission module automatically activates the backup satellite communication channel and transmits the encapsulated flood risk parameters to the pre-associated early warning execution module via the short message communication function of the BeiDou satellite navigation system. This method fully utilizes the advantages of BeiDou's short message service—wide coverage and independence from terrestrial networks—ensuring reliable communication in extreme weather or remote areas.

[0055] Conversely, when the wireless network signal strength is greater than or equal to the strength threshold, it indicates that the primary wireless communication link is in good condition. In this case, the data transmission module preferentially selects a higher-speed, lower-cost wireless network channel to transmit the flood risk parameters to the pre-associated early warning execution module.

[0056] This invention provides an intelligent dual-channel redundant transmission system that primarily uses a terrestrial wireless network and secondarily uses BeiDou short message communication. It can automatically select the optimal transmission path based on real-time channel conditions, thereby maximizing the reliability and timeliness of critical early warning information transmission under any environment.

[0057] According to the bridge flooding detection and prediction system provided by the present invention, both the warning light and the horn are closed-loop circuit control devices, which are activated only when the flooding risk exceeds a threshold; both the warning light and the horn are driven by a battery powered by a photovoltaic panel.

[0058] In this embodiment of the invention, the warning light and horn driven by the warning execution module are both highly reliable closed-loop circuit control devices. This closed-loop control circuit directly receives trigger commands from the warning execution module.

[0059] Under normal conditions, when the flood risk parameter does not exceed the risk threshold, the control circuit is in an open state, and the warning light and horn remain silent and do not consume any energy. Only when the flood risk parameter exceeds the preset risk threshold will the warning execution module send an activation signal to the control circuit, thereby closing the circuit loop and instantly activating the warning light to flash brightly and the horn to sound an alarm. This closed-loop control mechanism of normal openness and trigger closure greatly enhances the reliability and safety of the system, effectively avoids malfunctions, and reduces standby power consumption.

[0060] Furthermore, both the warning lights and the horn are driven by an independent power supply system, which consists of photovoltaic panels and a storage battery. Specifically, the photovoltaic panels are installed on the top of the light pole or in an unobstructed area around the bridge to convert solar energy into electrical energy. This electrical energy preferentially charges the connected storage battery, thereby storing the energy. The storage battery then provides all the necessary power to the closed-loop control circuits and actuators of the warning lights and horn.

[0061] Through the embodiments of the present invention, the power supply implementation method consisting of photovoltaic panels and batteries achieves energy self-sufficiency, enabling it to operate stably for a long time without relying on the bridge's power supply network. This feature is particularly suitable for bridges in remote areas or without electricity, ensuring that the local audible and visual alarm function can still be reliably activated and executed when the mains power is interrupted due to extreme weather, thereby greatly improving the robustness and independence of the entire early warning system in emergency situations.

[0062] According to the present invention, a bridge flooding sensing and prediction system is provided, wherein the system predicts the flooding risk parameters of downstream bridges based on the bridge state fusion parameters, combined with main stream hydrological data and tributary hydrological data, and through a hydrological dynamic evolution model, including: The bridge state fusion parameters, the main stream hydrological data, and the tributary hydrological data are input into the trained hydrological dynamic evolution model to perform dynamic evolution modeling and obtain the predicted value of downstream flooding time. The hydrological dynamic evolution model is used to correlate water level rise with flood propagation delay. The predicted downstream flooding time is quantified to generate flooding risk parameters that include the flooding range and evolution path.

[0063] In this embodiment of the invention, the process of predicting downstream bridge flooding risk parameters in the data processing module first involves data integration. The module aggregates and aligns the bridge state fusion parameters generated by the aforementioned multi-source sensing dynamic fusion model, the main stream hydrological data obtained in real-time from upstream hydrological stations, and the tributary hydrological data obtained from hydrological stations at the inlets of each tributary, forming a comprehensive model input dataset.

[0064] Subsequently, the integrated dataset is input into a pre-trained hydrological dynamic evolution model for dynamic evolution modeling. This model is a mathematical or intelligent algorithmic model capable of simulating the propagation process of flood waves in a river channel. Its core function lies in relating the complex nonlinear relationship between water level rise and flood propagation delay. Through the calculation of this model, the estimated time required for the flood to evolve from the current monitoring location to the downstream target bridge can be calculated, i.e., an accurate prediction of the downstream flooding time can be output.

[0065] Finally, the data processing module further quantifies the risk of the predicted downstream flooding time. This process goes beyond time prediction; it also considers current hydrological conditions, river topography, and bridge structural parameters to generate a final flooding risk parameter. This flooding risk parameter is a comprehensive risk indicator that includes not only the predicted flooding time but also the predicted inundation area (such as the area of ​​the bridge deck that may be submerged) and the flood's trajectory (such as the main direction of the flood flow).

[0066] In some embodiments, a suitable hydrological dynamic evolution model is selected, such as a data-driven machine learning model (e.g., Long Short-Term Memory Network (LSTM), Gated Recurrent Unit (GRU)). Hydrodynamic models can accurately describe the movement of water flow, but require a large number of physical parameters and complex calculations; machine learning models, on the other hand, can learn the relationship between water level rise and flood propagation delay from historical data, and have better adaptability and computational efficiency.

[0067] The preprocessed bridge state fusion parameters, main stream hydrological data, and tributary hydrological data are used as input features, and the actual downstream flooding time and flooding conditions are used as output labels to construct a training dataset. The dataset is then divided into a training set, a validation set, and a test set.

[0068] The selected model is trained using a training set, and its parameters (such as learning rate, number of iterations, number of network layers, etc.) are adjusted to minimize the loss function on the training set. During training, the model is validated using a validation set to prevent overfitting. Training is stopped when the model's performance on the validation set no longer improves. Finally, the trained model is evaluated using a test set to ensure it has good generalization ability.

[0069] The bridge status fusion parameters, main stream hydrological data, and tributary hydrological data collected in real time are preprocessed in the same way as the training data and then input into the trained hydrological dynamic evolution model.

[0070] The model dynamically evolves based on the input data and the correlation between water level rise and flood propagation delay. The hydrodynamic model simulates the propagation process of water flow in the river channel by solving hydrodynamic equations, predicting downstream water level changes and flooding time; the machine learning model extracts features and makes predictions on the input data based on patterns learned from historical data, and outputs a predicted value for downstream flooding time.

[0071] Based on the risk quantification results, and combined with Geographic Information System (GIS) technology, flood risk parameters including inundation extent and evolution path are generated. On the GIS platform, the predicted flooding time and water level information are overlaid with topographic data for analysis, and inundation extent maps and flood evolution path maps at different time points are drawn.

[0072] This invention comprehensively considers bridge condition fusion parameters, main stream hydrological data, and tributary hydrological data, taking into account various factors affecting downstream flooding. Bridge condition fusion parameters reflect the bridge's own flood resistance capacity, while main stream and tributary hydrological data provide information on flood source and propagation. By inputting this data into a trained hydrological dynamic evolution model, the flood propagation process can be simulated more accurately, downstream flooding time can be predicted, and the reliability of the prediction results can be improved.

[0073] According to the present invention, a bridge flooding sensing and prediction system is provided, wherein the bridge state fusion parameters are generated through a multi-source sensing dynamic fusion model based on the water level data, the geological displacement data, and the bridge deck water condition image data, including: A geological displacement risk coefficient is generated by dynamically correlating the water level data with the geological displacement data. The bridge's positioning data, geological offset risk coefficient, and bridge deck water condition image data are fused to obtain bridge state fusion parameters.

[0074] In this embodiment of the invention, the process of generating bridge state fusion parameters by the data processing module first involves a correlation analysis between water level and geological displacement. This module receives water level data and geological displacement data from the data acquisition module and performs time-series alignment and correlation calculations on these two types of data.

[0075] Specifically, the module's built-in algorithm analyzes the coupling relationship and hysteresis effect between rising (or falling) water levels and changes in bridge geological displacement. Based on this dynamic correlation analysis, a comprehensive geological displacement risk coefficient is generated. This coefficient quantifies the degree of risk of bridge foundation instability caused by changes in hydrogeological conditions.

[0076] Subsequently, the data processing module performs multi-source information fusion. The geological offset risk coefficient, the bridge's precise positioning data acquired via a GNSS receiver (i.e., its own geographic coordinates and elevation information), and bridge deck water condition image data captured by a camera (such as flood depth and extent) are all input into the multi-source sensing dynamic fusion model. This model employs specific data fusion algorithms (such as weighted fusion, Kalman filtering, or machine learning algorithms) to normalize and fuse these three types of heterogeneous data at the feature layer, ultimately outputting a bridge state fusion parameter that comprehensively reflects the bridge's current safety and flooding status.

[0077] Through the embodiments of the present invention, a deep fusion and collaborative perception of water level, geological displacement, bridge location and visual hydrological conditions are realized, transforming multiple single parameters into a more representative comprehensive indicator, providing an accurate and reliable data foundation for subsequent hydrological evolution prediction.

[0078] The following describes an example of the bridge flooding sensing and prediction system provided by the present invention in a practical application.

[0079] This invention mainly relies on data collection from front-end sensing stations. The equipment mainly includes: GNSS receiver, Beidou-3 data transmission terminal, camera, warning light, loudspeaker, industrial computer, water level sensor, etc. The sensing station integrates many devices and is mainly used for monitoring the water level of bridges and rivers to achieve the purpose of monitoring and early warning of bridge flooding.

[0080] The system can alert pedestrians and vehicles during and before flooding by flashing warning lights and blaring alarms, broadcasting announcements via loudspeaker, uploading water level data to the platform, and sending SMS messages to relevant personnel to monitor bridge flooding and prepare emergency plans. It can also notify nearby residents on-site. River flooding is common during the flood season or rainy season. In situations with unstable network conditions or without 4G / 5G network access, the system can use BeiDou short message technology to upload real-time water level information and send SMS messages to relevant personnel, ensuring no data loss.

[0081] If one or more bridges are located on the same river or tributary, or where the rivers intersect, the timing of downstream flooding can be predicted based on data such as upstream bridge flooding or water flow velocity. This allows for timely notification to downstream bridge managers or residents, enabling appropriate measures to be taken.

[0082] The receiver can upload high-precision positioning information to the platform in real time. It can monitor bridge displacement through positioning, predict flood disasters, and combine water level monitoring to achieve dual protection, ensuring accurate data transmission and accurate early warning.

[0083] refer to Figure 2 , Figure 2This is a hardware structure diagram of the bridge flooding perception and prediction system provided by the present invention, which includes: a Beidou-3 data transmission terminal, a loudspeaker, a camera (PTZ camera), a photovoltaic panel, a GNSS high-precision positioning and sensing terminal, a lightning rod, a camera (bullet camera), a warning light, and a control box.

[0084] Beidou-3 data transmission terminal: In extreme weather, there may be situations where there is no 4G / 5G signal. The Beidou-3 data transmission terminal can upload water level monitoring data via short message. In the event of flooding, it can notify the person in charge to carry out emergency response plan. Horn and warning lights: In the event of flooding, if the water level exceeds the warning threshold, the horn will be activated to issue a warning and the warning lights will sound to alert pedestrians, vehicles and other vehicles passing by. Camera (PTZ camera): Monitors the surrounding environment of the Beidou multi-source sensing station and has mobile monitoring capabilities; Camera (bullet camera): Monitors the bridge surface conditions. In flooded bridges where water level sensors are not suitable for installation, visual algorithms can be used to detect the water conditions and water flow rate on the bridge surface, achieving an auxiliary early warning effect. GNSS high-precision positioning and sensing terminal: monitors bridge displacement, can detect geological vibrations in advance when floods occur upstream, and provides early warnings through offset; Photovoltaic panels: primarily utilize solar energy to power storage batteries; Control box: Integrates photovoltaic controller, switch, and industrial edge computer. It can monitor various parameters of the photovoltaic controller, such as charging and discharging, voltage, etc., and monitor the power supply status of the equipment. The switch provides network access, mainly for cameras and GNSS high-precision positioning and sensing terminals. The industrial edge computer provides 4G / 5G network access and achieves data transmission and control of real-time camera / speaker operation via serial communication, achieving energy saving in non-powered states.

[0085] refer to Figure 3 , Figure 3 This is a flowchart of the BeiDou-3 short message data transmission process provided by the present invention.

[0086] The multi-source sensing terminal (i.e., the field terminal) sends data unidirectionally to the Beidou short message satellite via the Beidou data transmission terminal. After receiving the short message, the China Ordnance Beidou ground station forwards the short message to the central office via a dedicated fiber optic line. Finally, the short message is forwarded to the user platform via network communication.

[0087] Through this invention, firstly, a multi-source sensing dynamic fusion model is used to deeply integrate four types of heterogeneous data: satellite positioning, BeiDou short message service, water level sensor data, and visual perception data. This establishes a dynamic correlation between water level rise, geological displacement, and bridge deck water conditions, achieving centimeter-level precise sensing of water level and geological displacement. Secondly, by combining a meteorological hydrological dynamic evolution model and comprehensively utilizing hydrological data from the main stream and tributaries, the system accurately predicts the flooding time and trend of downstream bridges and generates a river flooding model prediction map. Finally, a dual-channel redundant transmission mechanism consisting of BeiDou short message service and 4G / 5G communication ensures stable transmission of critical early warning data under any network conditions, greatly improving the overall reliability of the system.

[0088] Compared with the existing technology, the technical solution proposed in this invention uses a closed-loop circuit for the warning light and horn to reduce energy consumption. In general, under extreme and rainy weather, there is little sunlight, and the photovoltaic panel cannot convert or converts only a small amount of electrical energy. In such weather, flooding is very likely to occur, and battery power supply is challenging. At the same time, the GNSS high-precision receiver has a lithium battery that can store electricity to maintain operation for a period of time. This solution can achieve energy saving to a certain extent. Furthermore, this invention can reduce the cost of manual on-site inspections and on-site warnings, reduce manpower output, and is highly adaptable, capable of adapting to the installation and monitoring of various types of flooded bridges; while having low construction costs, it also has high water level monitoring accuracy, reaching the centimeter level.

[0089] The bridge flooding sensing and prediction method provided by the present invention is described below. The bridge flooding sensing and prediction method described below can be referred to in correspondence with the bridge flooding sensing and prediction system described above.

[0090] refer to Figure 4 , Figure 4 This is a flowchart illustrating the bridge flooding sensing and prediction method provided by the present invention.

[0091] Step 401: Acquire multi-source monitoring data of the bridge in real time. The multi-source monitoring data includes: water level data collected by water level sensors, geological displacement data of the bridge monitored by GNSS receivers, and water condition image data of the bridge deck captured by cameras.

[0092] Step 402: Based on water level data, geological displacement data, and bridge deck water condition image data, bridge state fusion parameters are generated through a multi-source sensing dynamic fusion model.

[0093] Step 403: Based on the bridge status fusion parameters, combined with the hydrological data of the main stream and tributaries, the flood risk parameters of the downstream bridge are predicted through the hydrological dynamic evolution model.

[0094] Step 404: Trigger a graded response based on flood risk parameters: When the flood risk parameters exceed the risk threshold, execute a local audible and visual alarm via warning lights and horns, and simultaneously push remote early warning information to preset users via SMS platform.

[0095] Specifically, the bridge flooding sensing and prediction method provided by the present invention can realize all the method steps implemented in the above-mentioned bridge flooding sensing and prediction system embodiment, and can achieve the same technical effect. Here, the parts that are the same as those in the method embodiment and the beneficial effects will not be described in detail.

[0096] Figure 5 This is a schematic diagram of the physical structure of the electronic device provided by the present invention, such as... Figure 5 As shown, the electronic device may include: a processor 510, a communications interface 520, a memory 530, and a communications bus 540, wherein the processor 510, the communications interface 520, and the memory 530 communicate with each other through the communications bus 540. The processor 510 can call logic instructions in the memory 530 to execute a bridge flooding perception and prediction method. This method includes: acquiring multi-source monitoring data of the bridge in real time, wherein the multi-source monitoring data includes: water level data collected by a water level sensor, geological displacement data of the bridge monitored by a GNSS receiver, and bridge deck water condition image data captured by a camera; generating bridge state fusion parameters based on the water level data, the geological displacement data, and the bridge deck water condition image data through a multi-source perception dynamic fusion model; predicting downstream bridge flooding risk parameters based on the bridge state fusion parameters, combined with main stream hydrological data and tributary hydrological data, through a hydrological dynamic evolution model; and triggering a graded response according to the flooding risk parameters: when the flooding risk parameters exceed a risk threshold, executing a local audible and visual alarm through warning lights and a horn, and simultaneously pushing remote early warning information to preset users via an SMS platform.

[0097] Furthermore, the logical instructions in the aforementioned memory 530 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0098] On the other hand, the present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the bridge flooding perception and prediction method provided by the above methods. The method includes: acquiring multi-source monitoring data of the bridge in real time, wherein the multi-source monitoring data includes: water level data collected by a water level sensor, geological displacement data of the bridge monitored by a GNSS receiver, and bridge deck water condition image data captured by a camera; generating bridge state fusion parameters based on the water level data, the geological displacement data, and the bridge deck water condition image data through a multi-source perception dynamic fusion model; predicting downstream bridge flooding risk parameters based on the bridge state fusion parameters, combined with main stream hydrological data and tributary hydrological data, through a hydrological dynamic evolution model; and triggering a graded response according to the flooding risk parameters: when the flooding risk parameters exceed a risk threshold, executing a local audible and visual alarm through warning lights and a horn, and simultaneously pushing remote early warning information to preset users through an SMS platform.

[0099] In another aspect, the present invention also provides a non-transitory computer-readable storage medium storing a computer program thereon. When executed by a processor, the computer program implements the bridge flooding perception and prediction method provided by the above methods. The method includes: acquiring multi-source monitoring data of the bridge in real time, wherein the multi-source monitoring data includes: water level data collected by a water level sensor, geological displacement data of the bridge monitored by a GNSS receiver, and bridge deck water condition image data captured by a camera; generating bridge state fusion parameters based on the water level data, the geological displacement data, and the bridge deck water condition image data through a multi-source perception dynamic fusion model; predicting downstream bridge flooding risk parameters based on the bridge state fusion parameters, combined with main stream hydrological data and tributary hydrological data, through a hydrological dynamic evolution model; and triggering a graded response according to the flooding risk parameters: when the flooding risk parameters exceed a risk threshold, executing a local audible and visual alarm through warning lights and a horn, and simultaneously pushing remote early warning information to preset users through an SMS platform.

[0100] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0101] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0102] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A bridge flooding sensing and prediction system, characterized in that, include: The data acquisition module is used to acquire multi-source monitoring data of the bridge in real time. The multi-source monitoring data includes: water level data collected by a water level sensor, geological displacement data of the bridge monitored by a GNSS receiver, and bridge deck water condition image data captured by a camera. The data processing module is communicatively connected to the data acquisition module and is used to generate bridge state fusion parameters based on the water level data, the geological displacement data, and the bridge deck water condition image data through a multi-source sensing dynamic fusion model. Based on the bridge status fusion parameters, combined with the hydrological data of the main stream and the tributary, the flooding risk parameters of the downstream bridge are predicted through a hydrological dynamic evolution model. The data transmission module is communicatively connected to the data processing module and is used to transmit the flood risk parameters to the pre-associated early warning execution module through dual-channel redundancy. The early warning execution module is communicatively connected to the data transmission module and is used to trigger a graded response based on the flood risk parameter: when the flood risk parameter exceeds the risk threshold, a local audible and visual alarm is executed through warning lights and a horn, and at the same time, remote early warning information is pushed to preset users through the SMS platform.

2. The bridge flooding sensing and prediction system according to claim 1, characterized in that, The bridge surface water condition image data captured by the camera includes: A PTZ camera is used to dynamically monitor changes in the surrounding environment of the bridge; A bullet camera is used to identify the bridge deck water condition parameters through visual algorithms, wherein the bridge deck water condition parameters include flood depth and water flow velocity.

3. The bridge flooding sensing and prediction system according to claim 1, characterized in that, The step of transmitting the flood risk parameters to the pre-associated early warning execution module via dual-channel redundancy includes: Real-time monitoring of wireless network signal strength; When the wireless network signal strength is less than the strength threshold, the flood risk parameters are transmitted to the pre-associated early warning execution module via BeiDou short message; When the wireless network signal strength is greater than or equal to the strength threshold, the flood risk parameter is transmitted to the pre-associated early warning execution module via the wireless network.

4. The bridge flooding sensing and prediction system according to claim 1, characterized in that, Both the warning light and the horn are closed-loop circuit control devices, which are activated only when the risk of flooding exceeds the threshold; both the warning light and the horn are driven by batteries powered by photovoltaic panels.

5. The bridge flooding sensing and prediction system according to claim 1, characterized in that, The method of predicting downstream bridge flooding risk parameters based on the bridge state fusion parameters, combined with main stream and tributary hydrological data, using a hydrological dynamic evolution model includes: The bridge state fusion parameters, the main stream hydrological data, and the tributary hydrological data are input into the trained hydrological dynamic evolution model to perform dynamic evolution modeling and obtain the predicted value of downstream flooding time. The hydrological dynamic evolution model is used to correlate water level rise with flood propagation delay. The predicted downstream flooding time is quantified to generate flooding risk parameters that include the flooding range and evolution path.

6. The bridge flooding sensing and prediction system according to claim 1, characterized in that, The process of generating bridge state fusion parameters based on the water level data, the geological displacement data, and the bridge deck water condition image data through a multi-source sensing dynamic fusion model includes: A geological displacement risk coefficient is generated by dynamically correlating the water level data with the geological displacement data. The bridge's positioning data, geological offset risk coefficient, and bridge deck water condition image data are fused to obtain bridge state fusion parameters.

7. A method for sensing and predicting bridge flooding, characterized in that, include: Real-time acquisition of multi-source monitoring data of the bridge, wherein the multi-source monitoring data includes: water level data collected by water level sensors, geological displacement data of the bridge monitored by GNSS receivers, and bridge deck water condition image data captured by cameras; Based on the water level data, the geological displacement data, and the bridge deck water condition image data, bridge state fusion parameters are generated through a multi-source sensing dynamic fusion model. Based on the bridge status fusion parameters, combined with the hydrological data of the main stream and the tributary, the flooding risk parameters of the downstream bridge are predicted through a hydrological dynamic evolution model. A tiered response is triggered based on the flood risk parameters: when the flood risk parameters exceed the risk threshold, a local audible and visual alarm is triggered via warning lights and horns, and remote early warning information is pushed to preset users via SMS platform.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the bridge flooding perception and prediction method as described in claim 7.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the bridge flooding perception and prediction method as described in claim 7.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the bridge flooding perception and prediction method as described in claim 7.

Images