A channel regulation project dredging dynamic monitoring system and method

Abstract

The application relates to the technical field of waterway dredging, in particular to a waterway regulation engineering dredging dynamic monitoring system and method, which comprises a waterway monitoring module and a dredging operation monitoring module. The waterway monitoring module and the dredging operation monitoring module transmit data to an intelligent supervision module, the intelligent supervision module is connected with a construction execution terminal through a network, the waterway monitoring module is used for monitoring waterway topographic structure, hydrological conditions and ecological environment changes, the dredging operation monitoring module is used for monitoring equipment position, operation track, operation depth, dredging volume and construction picture monitoring, the intelligent supervision module receives data collected by the waterway monitoring module and the dredging operation monitoring module, and assists managers in timely regulating and controlling dredging operation. The waterway monitoring module and the dredging operation monitoring module are used for real-time dynamic feedback of dredging construction data and states, the intelligent supervision module receives data, and remote regulation and control are realized, so that whole-process fine dynamic management and control of dredging operation are realized.

Description

Technical Field

[0001] This invention relates to the field of waterway dredging technology, and in particular to a dynamic monitoring system and method for waterway improvement projects involving dredging. Background Technology

[0002] A waterway is a body of water in coastal areas, rivers, lakes, and canals where ships and rafts can navigate safely. It consists of three parts: navigable waters, navigational aids, and water conditions. Silt and sediment in waterways gradually clog the channels, threatening navigation safety. At the same time, pollutants accumulate in the waterways with the sediment year after year, continuously deteriorating the aquatic ecological environment. Therefore, regular dredging is necessary.

[0003] However, indiscriminate dredging not only wastes resources but also damages underwater topography and the ecological base. Therefore, systematic testing must be conducted before dredging. The existing monitoring system focuses on hydrodynamic parameters such as water depth, topography, and flow velocity, and dynamic monitoring focuses on navigation safety indicators. It lacks monitoring and control measures for the disturbance of aquatic ecology caused by dredging operations, which often causes serious damage to aquatic ecology after dredging.

[0004] On the other hand, there are shortcomings in intelligent supervision: First, there is a serious data silo problem among equipment. Data from multiple sources, such as drones, unmanned vessels, buoys, and shore-based stations, has not achieved true real-time fusion and joint decision-making, and still operates independently in a single-device, single-task mode. Second, AI applications are mainly for post-event analysis. AI in waterway dredging construction lacks predictive quality control capabilities and cannot predict in real time whether the dredging depth in a certain area will exceed the standard. Third, the "perception-decision-execution" chain has not yet been established. The construction process still relies heavily on human experience, and AI remains at the post-event analysis stage, not yet entering the real-time control stage. Summary of the Invention

[0005] The purpose of this invention is to address the shortcomings of existing technologies by proposing a dynamic monitoring system and method for dredging in waterway improvement projects.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: A dynamic monitoring system for dredging in waterway improvement projects includes a waterway monitoring module and a dredging operation monitoring module. These modules transmit data to an intelligent monitoring module, which is connected to a construction execution terminal via a network. The waterway monitoring module monitors waterway topography, hydrological conditions, and ecological environment changes. The dredging operation monitoring module monitors equipment location, operation trajectory, operation depth, and dredging volume, and monitors construction footage to prevent over-dredging, under-dredging, or excessive pollution. The intelligent monitoring module receives data collected by the waterway monitoring and dredging operation monitoring modules, completes the entire process of storage, processing, and analysis, and generates construction decision reports or issues operation instructions to assist management personnel in timely control of dredging operations. The intelligent monitoring module sends the construction decision reports or instructions to the construction execution terminal to drive the construction equipment, achieving refined dynamic control of the entire dredging operation process and completing precise dredging as needed.

[0007] Preferably, the channel monitoring module monitors the channel topography and structure, including monitoring the underwater topography and siltation of the channel, as well as monitoring the structural safety of fixed structures in the channel. The sensors used in the waterway monitoring module to monitor underwater topography and siltation include: multibeam sonar, side-scan sonar, high-frequency imaging sonar, microwave penetrating in-situ sensor, shallow seismic profiler, and dual-frequency acoustic siltation thickness measuring instrument.

[0008] Preferably, the equipment in the waterway monitoring module for monitoring the structural safety of fixed structures in the waterway includes: GNSS displacement stations are used to monitor the displacement and settlement of fixed structures in waterways. A piezometer is used to monitor the water pressure inside the soil and to calculate pore water pressure, the position of the phreatic line, and uplift pressure. An ultrasonic water depth sensor, combined with a GNSS displacement station and a piezometer, performs data fusion analysis to determine whether the structure is safe. Video surveillance equipment enables remote visual monitoring, anomaly warning, and data tracing.

[0009] Preferably, the monitoring of hydrological conditions includes monitoring water level, flow velocity, flow direction, and rainfall; The sensors or devices used to monitor hydrological conditions in the waterway monitoring module include: acoustic Doppler current profiler, radar current meter, hydrological buoy, rainfall sensor, and K-band planar radar flow meter.

[0010] Preferably, the monitoring of ecological environment changes includes monitoring of water quality and bottom sediment; The waterway monitoring module uses a multi-parameter water quality sensor to monitor water quality, including pH, dissolved oxygen, turbidity, conductivity, temperature, COD, and ammonia nitrogen. Suspended solids or heavy metals are rapidly sampled from river water or silt using a monitoring submersible, and then tested in a laboratory.

[0011] Preferably, the dredging operation monitoring module includes: GNSS or BeiDou positioning components are used to monitor the real-time location and navigation trajectory of the dredging vessel. The positioning components for the dredging robot include a shipborne RTK-GPS, a winch encoder, an A-frame tilt sensor, and an electric compass, which calculate the underwater three-dimensional coordinates of the dredging robot in real time based on multi-source data. High-frequency imaging sonar is used for construction monitoring in murky water, allowing for clear observation of the mud surface and obstacles, precise dredging, and acceptance of the thickness of residual mud after dredging. Draft sensors are used to monitor changes in a ship's load and estimate the amount of dredged material. AIS and ship radar are used for ship identification, collision avoidance, and preventing non-operational vessels from entering the dredging area.

[0012] Preferably, the intelligent monitoring module includes an edge computing node, a cloud server, a data storage device, and a display; the edge computing node connects the sensor and the cloud server, filters the data collected by the sensor, and uploads it to the cloud server; An AI inference engine and a digital twin engine are deployed on the cloud server. The digital twin engine integrates multi-source data in real time, calculates and generates a digital image of the river channel, and completes 3D scene rendering on the display for automated real-time panoramic monitoring. The AI ​​inference engine and the digital twin engine work together to drive intelligent analysis and risk warning, realizing a closed loop from data to decision-making, enabling intelligent control, automatic analysis, and automatic decision-making.

[0013] The method for dynamic monitoring of dredging in waterway improvement projects includes the following steps: S1. Based on the existing monitoring sensors and equipment in the waterway, reasonably deploy the remaining sensors and equipment that need to be fixedly installed in the waterway monitoring module to monitor the mud surface elevation, silt thickness, siltation rate, minimum navigable depth, flow velocity and direction, sediment content or sediment inflow, silt density or rheology, and water quality ecology for a long period of time. Organize historical data of the waterway to serve as the basis for dredging decisions; historical data includes hydrological data, sediment data, siltation data, dredging records, waterway cross-sections, meteorological data, and navigation data from previous years. S2. For sensors and equipment that require mobile monitoring in the waterway monitoring module, equip them with corresponding monitoring vessels and monitoring submersibles to carry out waterway inspection operations, collect full-section water depth topography, silt thickness distribution, cross-sectional flow velocity distribution, and water quality cross-section distribution, sample river water and bottom sediment, send them for testing and analysis, receive data at the edge computing node, complete data cleaning and screening locally, and upload core data to the cloud server. S3, the digital twin engine on the cloud server integrates a 3D model of the riverbed and underwater topographic point cloud, presents the siltation distribution and cross-sectional shape in real time, and uses sensor data to drive the twin to update in real time, perform panoramic mapping and dynamic simulation, simulate the effects of different dredging schemes, and select a scheme according to actual needs. S4, the AI ​​inference engine intelligently identifies the siltation range and predicts the dredging priority. The intelligent monitoring module uses big data and artificial intelligence technology to analyze and predict monitoring data. Through data models and algorithms, it identifies the risk of waterway siltation, predicts the dredging cycle and the priority of dredging work, and, combined with water level prediction, analyzes the siltation area at the corresponding water level and formulates the dredging tasks for each section. The S5 AI inference engine plans the construction path based on the dredging task, outputs decision reports or instructions, and after confirmation by management and construction personnel, the construction execution terminal controls the construction equipment to work. During the dredging process, the AI ​​inference engine automatically recommends the optimal parameters based on real-time working conditions, silt disturbance and diffusion, and water quality ecology. It also automatically optimizes and remotely controls the equipment. In case of problems such as sudden rise in water level, exceeding the dredging depth limit, soaring sand content, or abnormal equipment vibration, which personnel cannot react quickly or detect in time, the AI ​​inference engine will automatically alarm and control the equipment to stop or adjust the equipment operating parameters. For important decisions such as dredging plans or dredging schedules, the digital twin engine simulates multiple plans, and the AI ​​inference engine filters and compares the advantages and disadvantages of each plan before the final decision is made by the management. S6. Based on the dredging task, dredging is carried out in segments with precision to reduce damage to the ecology. The dredging equipment used includes dredging boats and dredging robots. The dredging boats are the main equipment for targeted and efficient dredging of the core siltation points with high dredging priority. The dredging robots are used as an auxiliary equipment to supplement the dredging of areas missed by the dredging boats, thereby expanding the dredging range. At the same time, they can carry out fine dredging of areas or thin mud layers that are inconvenient for dredging boats to reach. S7. Dredging data is transmitted back to the cloud server for centralized analysis and long-term storage, driving dynamic adjustments to the dredging strategy in real time. The cloud server also undertakes the AI ​​model training task. After the AI ​​model is trained, it is compressed and distributed to the edge computing node of the dredging robot to complete the model iteration and upgrade. S8. In the area where dredging has been completed, the monitoring vessel or monitoring submersible will conduct an inspection. The multi-source data collected by the waterway monitoring module will be integrated, and the dredging effect will be jointly verified by the AI ​​inference engine and the digital twin engine. The results will be determined to determine whether the contaminated layer is clean and whether any soil has been mistakenly excavated. The 3D models before and after dredging will be automatically compared to accurately calculate the dredging volume and water depth recovery. Meanwhile, the AI ​​inference engine learns the siltation pattern based on historical siltation data and hydrodynamic data, predicts the future siltation rate, automatically generates suggestions for the next dredging time, and performs predictive proactive dredging.

[0014] Preferably, in step S6, the dredging vessel and the dredging robot adopt a dual-platform collaborative mode, with different sensors and operating equipment configured on each dredging vessel or dredging robot as needed and according to budget, to achieve joint operation, reduce costs, and enrich functionality; the dredging vessel is equipped with a dredging robot positioning component to monitor the robot's position in real time and schedule the operation path of multiple machines. The dredging vessel is equipped with multibeam sonar to perform large-area scanning, supplemented by side-scan sonar to check underwater debris and detect siltation patterns, providing precise underwater sensing support for dredging operations; The dredging robot is equipped with high-frequency imaging sonar and uses real-time reasoning based on edge computing. It extracts acoustic image features through convolutional neural networks, identifies silt textures and obstacle shapes, and generates a three-dimensional real-scene topographic map. The system can autonomously perceive, plan paths, and precisely control depth to achieve precise dredging that only removes the contaminated layer and avoids obstacles.

[0015] Preferably, the monitoring submersible or dredging robot is also equipped with an attitude detection module, which includes an IMU inertial measurement unit, an electronic compass, a depth / pressure sensor, and a 9-axis attitude sensor. Combined with side-scan sonar and multi-beam sonar, it provides external reference and helps correct accumulated errors. An unmanned surface vessel is connected to the monitoring submersible or dredging robot. An electric winch is installed on the unmanned surface vessel, and a recovery rope is wound on the electric winch. One end of the recovery rope is connected to the upper side of the monitoring submersible or dredging robot, which facilitates the correction of the attitude of the monitoring submersible or dredging robot, or the rapid recovery of the monitoring submersible or dredging robot.

[0016] Compared with the prior art, the beneficial effects of the present invention are as follows: 1. This invention deploys an AI inference engine and a digital twin engine on a cloud server. The digital twin engine integrates multi-source data in real time, calculates and generates a digital image of the river channel, and completes 3D scene rendering on the display for automated real-time panoramic monitoring. At the same time, the AI ​​inference engine and the digital twin engine work together to drive intelligent analysis and risk warning, realizing a closed loop from data to decision-making, intelligent management and control, automatic analysis, automatic decision-making, data-driven, and predictive control. 2. This invention monitors changes in waterway topography, hydrological conditions, and ecological environment through a waterway monitoring module, and monitors equipment location, operation trajectory, operation depth, and dredging volume through a dredging operation monitoring module. It monitors the construction process and provides real-time dynamic feedback on dredging construction data and status. The intelligent supervision module receives the data, and the AI ​​inference engine automatically recommends optimal parameters based on real-time working conditions, silt disturbance and diffusion, and water quality ecology. It automatically optimizes and remotely controls the process, achieving refined and dynamic management of the entire dredging operation. It completes precise dredging as needed, removing only the surface polluted sludge while retaining the underlying native sludge, reducing ecological disturbance, significantly improving dredging efficiency, and making dredging operations more targeted. 3. This invention adopts a dredging strategy of "human control + intelligent unmanned equipment collaboration". Personnel drive monitoring vessels or dredging vessels to control the overall direction of monitoring and construction, and intervene in a timely manner in case of emergencies. At the same time, unmanned monitoring submersibles and unmanned dredging robots are deployed to make up for the shortcomings of dredging vessels and monitoring vessels, ensuring the controllability of the entire monitoring and dredging process. Personnel can also navigate vessels to avoid obstacles, and the monitoring submersibles and dredging robots can operate underwater to avoid affecting waterway traffic. In addition, when unmanned equipment malfunctions, personnel can quickly make adjustments or retrieve it to avoid equipment loss. Attached Figure Description

[0017] Figure 1 This is a control principle diagram of a dynamic monitoring system for dredging in waterway improvement projects proposed in this invention; Figure 2 This invention provides a flowchart of a dredging process for a waterway improvement project. Figure 3 This is a schematic diagram of the working status of an unmanned vessel for a dynamic monitoring system for dredging in waterway improvement projects proposed in this invention. Figure 4 This is a schematic diagram of the working status of the recovery buoy in a dynamic monitoring system for dredging in waterway improvement projects proposed in this invention. Figure 5 This is a schematic diagram of the working status of the monitoring vessel and monitoring submersible of the dynamic monitoring system for dredging in waterway improvement projects proposed in this invention.

[0018] In the picture: 1. Dredging vessel; 2. Dredging robot; 3. Unmanned vessel; 4. Monitoring vessel; 5. Monitoring submersible. Detailed Implementation

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The following description of at least one exemplary embodiment is merely illustrative and is in no way intended to limit the present invention or its application or use. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] Reference Figure 1-5 A dynamic monitoring system for dredging in waterway improvement projects includes a waterway monitoring module and a dredging operation monitoring module. The waterway monitoring module and the dredging operation monitoring module transmit data to an intelligent supervision module, which connects to the construction execution terminal via a network. The waterway monitoring module is used to monitor changes in waterway topography, hydrological conditions, and ecological environment; the dredging operation monitoring module is used to monitor equipment location, operation trajectory, operation depth, and dredging volume to prevent over-dredging, under-dredging, or excessive construction pollution. The intelligent monitoring module receives data collected by the waterway monitoring module and the dredging operation monitoring module, completes the entire process of storage, processing and analysis, and generates construction decision reports or issues operation instructions based on the data to assist managers in timely control of dredging operations. The intelligent monitoring module sends the construction decision reports or instructions to the construction execution terminal to drive the construction equipment to work, realize refined and dynamic control of the entire dredging operation process, complete precise dredging as needed, and reduce the impact on the ecological environment.

[0021] The channel monitoring module monitors the channel topography and structure, including underwater topography and siltation, as well as the structural safety of fixed structures in the channel. The sensors used in the waterway monitoring module to monitor underwater topography and siltation in the waterway include: Multibeam sonar is used to scan the riverbed topography and silt thickness distribution over a large area along the entire line, generating a three-dimensional topographic model and accurate water depth map; it is used to assist in calculating earthwork volume before dredging, guide dredging depth during dredging, and re-measure water depth after dredging to confirm that the design elevation has been reached; multibeam sonar can be used for periodic rescanning as needed, and the siltation rate can also be statistically analyzed. Side-scan sonar can quickly scan the entire line over a large area to find obstacles, shallows, and hard bottoms. Before construction, it can also be used to survey the underwater debris (sunken ships, rocks, and pipe piles) and detect the siltation patterns. Multibeam sonar is characterized by lower resolution, lower signal-to-noise ratio, and high positioning accuracy, while side-scan sonar has higher resolution, higher signal-to-noise ratio, and lower positioning accuracy. The two complement each other and are used for preliminary surveys before construction. High-frequency imaging sonar uses sound waves instead of light to form images. Its frequency is usually 500kHz to 1.2MHz+, and its resolution is at the millimeter level. However, its effective range is short (within tens of meters). It is used for real-time monitoring in turbid water construction, to check the thickness of residual mud after dredging, and can also be used to check for cracks in bank protection, identify fish species, and count their numbers. Microwave penetrating in-situ sensor uses 1-10GHz microwaves to penetrate the surface of silt and invert the moisture content. Without sampling, it can know how wet the silt is and how many layers it is. Before construction, it can detect the bottom silt and determine the end point of dredging. The probe can also be permanently buried for long-term in-situ monitoring, real-time reading of moisture content changes, and tracking of silt return rate. The shallow seismic profiler uses low-frequency acoustic waves to obtain sufficient penetration power to detect the distribution of silt and the stratigraphic structure. Before dredging, it detects the silt layer structure; after dredging, it re-scans the profile to confirm whether the dredging has reached the designed depth; it can also be re-scanned periodically to facilitate monitoring of the silt return rate and stratigraphic changes. Dual-frequency acoustic silt thickness measuring instrument: It emits two sound waves of high and low frequencies at the same time. The high frequency is bounced back by the silt surface, while the low frequency penetrates the silt and bounces back from the hard bottom. The difference between the two water depths is the silt thickness; with an accuracy of ±5cm, it can scan the entire silt map during mobile inspection and is the core equipment for pre-dredging assessment and acceptance comparison. A dual-frequency acoustic sediment thickness measuring instrument measures the thickness of the sediment layer, a shallow seismic profiler penetrates the strata to obtain the layered structure, and a water quality sensor collects pollution indicators. The three sources of data are fused and analyzed by AI to automatically determine whether the current area is a polluted layer or good soil.

[0022] The equipment used in the waterway monitoring module to monitor the structural safety of fixed structures in the waterway includes: GNSS displacement stations are used to monitor the displacement and settlement of fixed structures in waterways, such as: horizontal displacement and settlement of dikes and revetments; dredging and hollowing out the dike toe can lead to overall slippage, with horizontal slippage >5mm / d triggering a yellow alert; 10mm / d triggering a red alert; horizontal displacement and tilting of groynes and bottom revetments; dredging alters water flow, leading to scour instability, and sudden changes in displacement rate >3mm / h triggering an immediate alarm; three-dimensional displacement of locks and bridge piers; dredging disturbs the foundation, leading to uneven settlement, with settlement >2mm / d triggering a yellow alert; creep displacement of waterway slopes; dredging alters slope stress, leading to slow landslides, with cumulative displacement >30mm triggering a red alert; surface slippage of slopes in siltation areas; dredging exposes silt, leading to slope instability, with cumulative displacement >30mm triggering a red alert. It can also use 3D laser scanning to fix the structure of buildings in the waterway, and perform full-coverage scanning to obtain a 3D point cloud model of the building structure. It can measure complex shapes, has strong visualization, and can directly output 3D models. Deformation areas can be viewed on a large screen, and historical data can be compared. After scanning at least twice, point cloud comparison can be performed to accurately calculate structural changes with an accuracy of up to millimeters. Piezometers can be permanently installed in dikes, groynes, and sluice gate foundations. Vibrating wire piezometers have no electronic components, have a long lifespan, and are resistant to interference. They are used to monitor the water pressure inside the soil, calculate pore water pressure, the position of the phreatic line, and uplift pressure. Seepage damage is often insidious and gradual, making it difficult to detect with the naked eye. Piezometers can collect data 24 hours a day, accurately detect abnormal pressure fluctuations, and issue early warnings at the initial stage of danger, preventing major accidents such as dam failure, dike collapse, and landslides from the source. Ultrasonic water depth sensors are used to monitor the difference in water depth before and after dredging, determine whether the design depth has been reached, and can also periodically monitor water depth changes to calculate siltation and scouring rates. When linked with a water level gauge, they can also calculate mud surface elevation, siltation return rate, dredging effect, and minimum water depth of the channel. Water depth is fundamental data for monitoring the structural safety of fixed structures in waterways. In conjunction with GNSS displacement stations and piezometers, data fusion analysis can determine whether a structure is safe. For example, a sudden change in water depth, such as a change of more than 0.5m in a short period of time, triggers an early warning. A sudden drop in water depth may indicate that the riverbed is collapsing, which is a precursor to structural instability. As needed, video surveillance equipment can also be deployed on fixed buildings or construction sites to achieve remote visual monitoring, anomaly warning, data traceability, and parameter debugging. The video surveillance equipment can effectively monitor fixed buildings or construction sites remotely and detect anomalies in a timely manner. The video surveillance equipment can also upload the collected video image data to the intelligent supervision module via wired or wireless network, and store the data through data storage devices to facilitate later work quality traceability or accident cause investigation.

[0023] Monitoring of hydrological conditions includes monitoring water level, flow velocity, flow direction, and rainfall; The sensors or devices used to monitor hydrological conditions in the waterway monitoring module include: Acoustic Doppler velocity profiler: acquires vertical velocity in real time and monitors flow field changes; full-section velocity scanning before dredging of large rivers to monitor cross-sectional velocity distribution. A radar current meter, fixed on the bank, measures the surface velocity of water. By analyzing changes in velocity in the construction area, it assesses the extent of suspended solids diffusion and its impact on downstream water intakes. Hydrological buoys continuously monitor water level, flow velocity, and water quality, with multiple parameters monitored simultaneously. When using water quality buoy stations, integrated monitoring of six parameters—water temperature, conductivity, pH, dissolved oxygen, ammonia nitrogen, and turbidity—can also be achieved. Rainfall sensors monitor rainfall, provide early warnings of heavy rain, suspend dredging operations, and schedule dredging windows. In addition, K-band planar radar flowmeters can be used to measure water level, flow velocity, flow rate, and rainfall in a non-contact manner, combining these four parameters to calculate the changes in the river's flow capacity before and after dredging and to determine whether the dredging was effective.

[0024] Monitoring changes in the ecological environment includes monitoring water quality and bottom sediment. After dredging, regular monitoring of aquatic ecology is also required. When bottom sediment is turned up during dredging, it will release a large amount of pollutants instantly, which requires timely monitoring and control of the construction progress to avoid serious damage to the river's ecological environment. Water quality monitoring items include turbidity, suspended solids, pH, dissolved oxygen, and heavy metals; The waterway monitoring module uses a multi-parameter water quality sensor to monitor water quality. The multi-parameter water quality sensor can monitor pH, dissolved oxygen, turbidity, conductivity, temperature, COD, and ammonia nitrogen. When waterway dredging causes disturbance and leads to a deterioration in water quality, the multi-parameter water quality sensor will automatically alarm. After dredging, it is used to monitor whether the water quality has recovered. Multispectral high-definition cameras can be used to capture the distribution and diffusion boundaries of turbidity on the water surface, while lidar can determine the level of turbidity. Heavy metal concentrations can be determined using an online rapid heavy metal detector or an atomic fluorescence spectrometer. Detailed analysis of ammonia nitrogen or heavy metals requires sampling. Laboratory testing can be performed by using a monitoring submersible to quickly sample river water or silt. The monitoring submersible uses a vacuum negative pressure bottle for sampling. The bottle opening is controlled by a solenoid valve. After reaching the sampling position, the solenoid valve opens, and the negative pressure inside the vacuum negative pressure bottle is used to quickly complete the sampling of river water or bottom sediment. Vacuum negative pressure bottle sampling is only suitable for rough chemical composition analysis. It is efficient, fast, and can be used in conjunction with unmanned monitoring boats and unmanned detection submersibles. The monitoring vessel uses an onboard miniature columnar sampler to take complete samples of the bottom sediment, obtaining undisturbed columnar sediment samples, which are then sent to the laboratory for mechanical testing or microstructure analysis.

[0025] The dredging operation monitoring module includes: GNSS or BeiDou positioning components are used to monitor the real-time location and navigation trajectory of the dredging vessel. The positioning component for the dredging robot utilizes a towed cable for positioning, an onboard RTK-GPS to determine the dredging vessel's position, a winch encoder to measure the towed cable's extended length, an A-frame tilt sensor and gyrocompass to measure the towed cable's entry angle and azimuth, and a depth sensor to determine the water depth, eliminating entry angle errors. Based on multi-source data from the onboard RTK-GPS, winch encoder, A-frame tilt sensor, and gyrocompass, the underwater three-dimensional coordinates of the dredging robot are calculated in real time with an accuracy better than ±0.3m and an update frequency of 10Hz, meeting the positioning accuracy and real-time requirements of waterway dredging operations. The cableless robot uses an underwater acoustic positioning system, and the dredging vessel dispatch instructions are issued through acoustic communication. The dredging robot calculates its own position using integrated navigation. The two complement each other and are dispatched in real time. The underwater acoustic positioning system includes a USBL ultra-short baseline, a Doppler velocimeter, and inertial navigation. The dredging vessel is equipped with a USBL ultra-short baseline at the bottom to measure the absolute position, with correction every second and an accuracy of ±0.5m to ±2m. The cableless robot is equipped with a Doppler velocimeter and inertial navigation; the Doppler velocimeter detects high-precision relative motion; the inertial navigation uses accelerometers and gyroscopes to calculate position and fill in the blind spots of USBL.

[0026] High-frequency imaging sonar is used for monitoring during muddy water construction, clearly seeing the mud surface and obstacles, and accurately clearing silt; the thickness of residual mud is inspected after siltation. Draft sensors are used to monitor changes in the ship's load and estimate the dredging volume, eliminating the need to monitor the cutter head status and the flow rate in the dredging pipe, thus reducing costs. As needed, a mud pump flow meter can be configured to further accurately monitor the mud delivery volume and perform real-time volume statistics; The dredging vessel and monitoring vessel are equipped with AIS and ship radar. The AIS broadcasts the basic information, location and operation status of the dredging vessel, and surrounding vessels automatically receive the information to avoid collisions. The AIS can also broadcast virtual navigation marks to inform surrounding vessels of the boundaries of the dredging operation area and remind them to give way. The ship radar is used for close-range obstacle avoidance. While the dredging vessel is operating in the waterway, the radar monitors surrounding vessels, buoys and shorelines in real time. GNSS or Beidou positioning components provide AIS and radar with the ship's precise location. AIS identifies the ship, and the ship's radar calculates distance and bearing, ensuring that the dredging vessel can operate safely in the waterway around the clock and under all working conditions.

[0027] The intelligent monitoring module includes edge computing nodes, cloud servers, data storage devices, and displays; Edge computing nodes connect sensors and cloud servers, completing data cleaning, filtering, and real-time control locally, and uploading only high-value data to the cloud server; Deploy AI inference engine and digital twin engine on cloud server. The AI ​​inference engine loads models such as siltation risk prediction model, structural safety early warning model, dredging volume calculation model, scheduling decision model, and water pollution diffusion prediction model. By fusing multi-source data and dynamic spatiotemporal mapping, a 3D mirror of the river is constructed in the digital twin engine to simulate and predict the dredging range under different water levels. For example, by changing the water level, we can predict changes in the dredging area; by changing the cutter head speed, we can predict the dredging efficiency; and by changing the rainfall, we can predict the siltation backfilling speed. The digital twin engine integrates multi-source data in real time, calculates and generates a digital mirror of the river channel, and completes 3D scene rendering on the display, enabling automated real-time panoramic monitoring and multi-level early warning. For example, side-scan topography, ADCP water flow, cutterhead status, AIS vessels, and water quality parameters can all be overlaid in a 3D river channel; when anomalies occur, such as pipe blockage, overloading, deviation from the channel, or illegal intrusion, the digital twin engine will automatically identify and push an alarm, without the need for manual supervision; Based on this, the AI ​​inference engine and the digital twin engine work together to drive intelligent analysis and risk warning, realizing a closed loop from data to decision-making, intelligent control, automatic analysis, and automatic decision-making; The dredging vessel is equipped with a ship-side edge computing node, on which a ship-side AI is deployed to schedule multiple dredging robots to work collaboratively. During dredging, multiple dredging vessels and robots operate simultaneously. The AI ​​on the vessel assigns dredging areas to the robots, plans routes, and avoids collisions. After the dredging vessels have finished dredging, the remaining silt is further cleaned by the dredging robots. When the dredging vessels are difficult to adjust, the dredging robots are used to supplement them, improving dredging efficiency. The dredging vessels are responsible for dredging large areas with severe siltation, as well as filtering, compacting, storing, and transporting the silt, while also discharging the filtered water. The ship's AI divides the dredging area equally among the dredging robots according to the thickness of the silt and the degree of pollution. During the dredging process, if the dredging is too fast or too slow in a certain area, the dredging robots are scheduled in real time to improve work efficiency and avoid collisions between dredging robots or entanglement of tow cables. After the dredging operation is completed, the digital twin engine retrieves data monitored by the waterway monitoring module for comparison, and evaluates the changes in water level and flow velocity under the same flow rate before and after dredging.

[0028] The dynamic monitoring method includes the following steps: S1. Based on the existing monitoring sensors and equipment in the waterway, reasonably deploy the remaining sensors and equipment that need to be fixedly installed in the waterway monitoring module to monitor the mud surface elevation, silt thickness, siltation rate, minimum navigable depth, flow velocity and direction, sediment content or sediment inflow, silt density or rheology, and water quality ecology for a long period of time. Organize historical data of the waterway to serve as the basis for dredging decisions; historical data includes hydrological data, sediment data, siltation data, dredging records, waterway cross-sections, meteorological data, and navigation data from previous years. S2. For sensors and equipment that require mobile monitoring in the waterway monitoring module, equip them with corresponding monitoring vessels and monitoring submersibles to carry out waterway inspection operations, collect full-section water depth topography, silt thickness distribution, cross-sectional flow velocity distribution, and water quality cross-section distribution, sample river water and bottom sediment, send them for testing and analysis, receive data at the edge computing node, complete data cleaning and screening locally, and upload core data to the cloud server. S3, the digital twin engine on the cloud server integrates the three-dimensional model of the riverbed and the underwater topographic point cloud, presenting the siltation distribution and cross-sectional shape in real time. Sensor data drives the twin to update in real time, panoramic mapping, and the physical world and digital mirror are synchronized in seconds. The siltation thickness, mud surface elevation and minimum water depth are clear at a glance. Dynamic simulations are used to model the effects of different dredging schemes—which section to dredge, how deep to dredge, and how to drain the silt. The schemes are tested before construction, and different schemes are simulated to select the appropriate scheme based on actual needs. S4, the AI ​​inference engine intelligently identifies the siltation range and predicts the dredging priority. The intelligent monitoring module uses big data and artificial intelligence technology to analyze and predict monitoring data. Through data models and algorithms, it identifies the risk of waterway siltation, predicts the dredging cycle and the priority of dredging work, and, combined with water level prediction, analyzes the siltation area at the corresponding water level and formulates the dredging tasks for each section. The S5 AI inference engine plans the construction path based on the dredging task, outputs decision reports or instructions, and after confirmation by management and construction personnel, the construction execution terminal controls the construction equipment to work. During the dredging process, the AI ​​inference engine automatically recommends the optimal parameters and optimizes them remotely based on real-time working conditions, silt disturbance and diffusion, and water quality ecology. For example, if there are problems that personnel cannot react quickly or detect in time, such as sudden rise in water level, exceeding the dredging depth limit, soaring sand content, or abnormal equipment vibration, the AI ​​inference engine will automatically alarm and control the equipment to stop or adjust the equipment operating parameters. For important decisions such as dredging plans or dredging schedules, the digital twin engine simulates multiple plans, and the AI ​​inference engine filters and compares the advantages and disadvantages of each plan before the final decision is made by the management. For example, the amount of dredging is used as the result indicator, and the amount of silt disturbance is used as the process control indicator. Real-time control indicators during construction include: Ground Loss Rate (GLR), GLR = ΔV / V × 100%, is the proportion of soil loss per unit volume; ΔV is the volume of soil lost due to disturbance, and V is the original soil volume; When GLR < 2‰ (low disturbance), the control strategy is to dredge at full speed to maximize efficiency; For a disturbance of 2‰≤GLR<5‰, the control strategy is to reduce speed and blade pressure, and adopt fuzzy PID adaptive adjustment, which is the mainstream solution that has been verified in current underwater dredging robots. In view of the strong nonlinearity and time-varying nature of the underwater environment, a neural network compensation algorithm is introduced to learn and correct the residual of the fuzzy PID, which significantly improves the control stability. GLR≥5‰ (exceeding the limit), the control strategy is pause / rollback, switch to low disturbance mode; Disturbance radius: the range in which the soil undergoes plastic deformation, centered on the dredging tool. Surface settlement, cumulative settlement of the construction section; The increase in pore water pressure, Δu=u 扰动 -u 静水 The excess pore water pressure caused by dredging is measured using a vibrating wire pore water pressure gauge or a piezoresistive pore water pressure gauge, and is used as a real-time proxy indicator of the disturbance intensity. The sensor must be 2 to 5 meters upstream of the dredging operation surface; otherwise, the measured value will be the value after the disturbance, not the baseline value before the disturbance.

[0029] Using the degree of dispersion of pollutants in sludge as a control indicator, the Dispersion Index (DPI) is defined as follows: , Suspended matter diffusion term SS represents real-time turbidity / suspended solids concentration, read from an online turbidity meter, in NTU or mg / L. limit The threshold is determined based on the water area's sensitivity level. Heavy metal diffusion term C metal Real-time heavy metal concentrations (maximum single value), including Pb, Cd, Hg, As, and Cr(VI), can be obtained using an online rapid heavy metal detector or atomic fluorescence spectrometry, in mg / L (aqueous phase) or mg / kg (dry weight of sediment). limit Corresponding national standard limits; Physical diffusion radius term R diff The distance from the leading edge of the pollution plume to the dredging section in real time is obtained from multiple monitoring wells, in meters (m) and radius (R). limit Control thresholds, for example, strictly control the depth of drinking water sources to 3m, generally control the depth of urban rivers to 5m, and control the depth of general water areas to 10m. Pore ​​water pressure driving term , Incremental pressure of excess pore water, unit kPa, pore water pressure gauge, real-time sampling. Control threshold; It is a bridging variable connecting disturbance and diffusion; dredging disturbance generates excess pore water pressure. The effective stress decreases, the soil becomes unstable, resuspension increases, and diffusion and exchange are enhanced. Weighting:

[0030] Weighting design principle: Give the highest weight to the most dangerous substances; heavy metals are the most deadly in drinking water sources. =0.40; Pore water pressure in farmland and waterways is the most critical factor (fluidization risk). =0.40; By controlling the diffusion index (DPI), dredging operations can be prevented from causing serious pollution and pollution from dredging operations can be effectively reduced. For example, the monitoring data for urban river dredging is as follows:

[0031] Substitute into the formula (urban river channel weight): DPI=0.25×1.80+0.30×1.20+0.20×0.80+0.25×0.80=1.17 Judgment: DPI=1.17>1.0, exceeding the limit, immediately suspend dredging; S6. Based on the dredging task, dredging will be carried out in segments with precise timing and rotation to allow time for ecological recovery. Dredging will be precisely targeted at areas with severe siltation to minimize ecological damage. Dredging equipment will include dredging vessels and dredging robots. Environmentally friendly trailing suction hopper dredgers or spiral suction dredgers will be used, with spiral suction dredgers equipped with environmentally friendly cutter heads. Closed-loop conveying minimizes silt spread, reduces disturbance to the original soil, achieves high bottom silt removal rates, and allows for precise depth control, preventing over-excavation. Dredging vessels will be the primary means of targeted and efficient dredging of high-priority core siltation points. Dredging robots will be used as a supplementary means to dredge areas missed by the dredging vessels and to perform fine dredging of areas inaccessible by dredging vessels or thin silt layers. S7. Dredging data is transmitted back to the cloud server for centralized analysis and long-term storage, driving dynamic adjustments to the dredging strategy in real time; local communication is supported between dredging robots to achieve multi-robot collaborative operation; the cloud server also undertakes AI model training tasks, and after training, the data is compressed and distributed to the edge computing nodes of the dredging robots to complete model iteration and upgrade; S8. In the area where dredging has been completed, the monitoring vessel or monitoring submersible will conduct an inspection. The multi-source data collected by the waterway monitoring module will be integrated, and the dredging effect will be jointly verified by the AI ​​inference engine and the digital twin engine. The results will be determined to determine whether the contaminated layer is clean and whether any soil has been mistakenly excavated. The 3D models before and after dredging will be automatically compared to accurately calculate the dredging volume and water depth recovery. Meanwhile, the AI ​​inference engine learns the siltation pattern based on historical siltation data and hydrodynamic data, predicts the future siltation rate, automatically generates suggestions for the next dredging time, and performs predictive proactive dredging to avoid affecting the waterway before dredging.

[0032] In addition, in the S6, the dredging vessel and the dredging robot adopt a dual-platform collaborative mode. Each dredging vessel or dredging robot is equipped with different sensors and operating equipment as needed and according to the budget to achieve joint operation. The dredging vessel is equipped with a dredging robot positioning component, which monitors the robot's position in real time and schedules the operation path of multiple robots. The shipboard AI scheduling engine, based on a digital twin map, divides the work area into several sub-zones according to siltation thickness and pollution levels, assigning them to various robots for collaborative scheduling. In the cabled solution, the cable itself constitutes a physical constraint, naturally preventing collisions between robots; the scheduling system only needs to prevent cable entanglement.

[0033] The dredging vessel is equipped with multibeam sonar to perform large-area scanning, supplemented by side-scan sonar to check underwater debris and detect siltation patterns, providing precise underwater sensing support for dredging operations; The dredging robot is equipped with high-frequency imaging sonar and uses real-time reasoning based on edge computing. It extracts acoustic image features through convolutional neural networks, identifies silt textures and obstacle shapes, and generates a three-dimensional real-scene topographic map. The system can autonomously perceive, plan paths, and precisely control depth to achieve precise dredging that only removes the contaminated layer and avoids obstacles.

[0034] The monitoring submersible or dredging robot is also equipped with an attitude detection module, which includes an IMU inertial measurement unit, an electronic compass, a depth / pressure sensor and a 9-axis attitude sensor. Combined with side-scan sonar and multi-beam sonar, it provides external reference and helps to correct accumulated errors. The attitude detection module can be selected and used according to usage requirements and cost budget; the attitude monitoring of the submersible's underwater navigation relies on the collaboration of multiple sensors: the IMU senses the angular velocity, the compass locks the heading, the depth gauge constrains the longitudinal position, and the sonar / vision provides external calibration; the data from each channel are fused by Kalman filtering or complementary filtering to output real-time three-axis attitude estimation, and then the closed-loop control system drives the thruster to complete the deviation correction; The IMU (Inertial Measurement Unit), including a three-axis gyroscope and a three-axis accelerometer, senses the angular velocity and acceleration in three dimensions: roll, pitch, and yaw in real time, enabling the monitoring of the attitude perception of submersibles or dredging robots. An electronic compass (magnetometer) monitors the heading angle and is used to correct heading drift caused by long-term operation of the gyroscope, ensuring that the underwater robot does not deviate from its course. Depth / pressure sensors monitor the longitudinal water depth of underwater robots, facilitating the monitoring of the submersible's lock depth and resisting vertical convection disturbances in the water. The dredging robot can be individually configured with a 9-axis attitude sensor, a 3-axis accelerometer, a 3-axis gyroscope, and a 3-axis magnetometer. Each of the three sets of sensors outputs 3 channels, providing a total of 9 dimensions of motion data. After fusion, the complete three-dimensional attitude (roll, pitch, yaw) is output. 9-axis is the minimum complete configuration for underwater attitude monitoring.

[0035] Monitoring submersibles can actively correct their attitude through thrusters, but dredging robots generally use tracks, which cannot complete attitude correction and require manual intervention. Therefore, an unmanned vessel is connected to the monitoring submersible and the dredging robot. An electric winch is installed on the unmanned vessel, and a recovery rope is wound on the electric winch. One end of the recovery rope is connected to the monitoring submersible and the dredging robot. In the complex underwater environment of waterways, in order to quickly recover or adjust the position and attitude of monitoring submersibles and dredging robots, the electric winch on the unmanned vessel can be used to wind up the recovery rope, thereby using the buoyancy of the unmanned vessel to lift the monitoring submersible and dredging robot, complete obstacle avoidance or rapid movement, and avoid direct dragging, which could cause equipment damage. Meanwhile, when the monitoring submersible and dredging robot are restricted by debris at the bottom of the channel, the thruster of the monitoring submersible may become blocked or entangled by debris, affecting the thruster's ability to drive the monitoring submersible. The dredging robot is generally tracked. If it capsizes underwater and is difficult to adjust, it needs to be pulled and adjusted by an unmanned boat using a retrieval rope to quickly adjust the dredging robot's attitude. The unmanned boat can also assist in locating the monitoring submersible and dredging robot. The unmanned vessel can also be replaced with a recovery buoy. The recovery buoy is brightly colored and can also be equipped with a locator as needed for quick positioning on large water surfaces. The monitoring submersible and dredging robot are equipped with a winding reel. The rotating part of the winding reel is fixed by an electromagnet. A connecting line is wound around the winding reel, and one end of the connecting line is connected to the recovery float. In normal operation, the connecting line is wound around the reel, and the retrieval buoy is located on top of the monitoring submersible and the dredging robot. When the equipment malfunctions, the power supply to the electromagnet is cut off, or when the equipment malfunction causes a power outage, the retrieval buoy drags the connecting line to the surface, making it easy for maintenance personnel to quickly locate and retrieve the monitoring submersible and the dredging robot.

[0036] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

[0037] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the exemplary embodiments according to this application. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0038] Unless otherwise specifically stated, the relative arrangement, numerical expressions, and values ​​of the components and steps set forth in these embodiments do not limit the scope of the invention. It should also be understood that, for ease of description, the dimensions of the various parts shown in the drawings are not drawn to actual scale. Techniques, methods, and devices known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods, and devices should be considered part of the specification. In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limitations. Therefore, other examples of exemplary embodiments may have different values. It should be noted that similar reference numerals and letters in the following figures denote similar items; therefore, once an item is defined in one figure, it need not be further discussed in subsequent figures.

Claims

1. A dynamic monitoring system for dredging in waterway improvement projects, characterized in that, It includes a waterway monitoring module and a dredging operation monitoring module, which transmit data to an intelligent supervision module, which is connected to the construction execution terminal via a network; The waterway monitoring module is used to monitor changes in waterway topography, hydrological conditions, and ecological environment; the dredging operation monitoring module is used to monitor equipment location, operation trajectory, operation depth, and dredging volume, and to monitor construction footage to prevent over-dredging, under-dredging, or excessive construction pollution. The intelligent monitoring module receives data collected by the waterway monitoring module and the dredging operation monitoring module, completes the entire process of storage, processing and analysis, and generates construction decision reports or issues operation instructions accordingly to assist managers in timely control of dredging operations. The intelligent monitoring module sends the construction decision reports or instructions to the construction execution terminal to drive the construction equipment to work, realizes refined and dynamic control of the entire dredging operation process, and completes precise dredging as needed.

2. The dynamic monitoring system for dredging in waterway improvement projects according to claim 1, characterized in that, The channel monitoring module monitors the channel topography and structure, including underwater topography and siltation, as well as the structural safety of fixed structures in the channel. The sensors used in the waterway monitoring module to monitor underwater topography and siltation include: multibeam sonar, side-scan sonar, high-frequency imaging sonar, microwave penetrating in-situ sensor, shallow seismic profiler, and dual-frequency acoustic siltation thickness measuring instrument.

3. The dynamic monitoring system for dredging in waterway improvement projects according to claim 2, characterized in that, The equipment used in the waterway monitoring module to monitor the structural safety of fixed structures in the waterway includes: a GNSS displacement station, a piezometer, an ultrasonic depth sensor, and video surveillance equipment.

4. The dynamic monitoring system for dredging in waterway improvement projects according to claim 1, characterized in that, The monitoring of the hydrological conditions includes the monitoring of water level, flow velocity, flow direction and rainfall; The sensors or devices used to monitor hydrological conditions in the waterway monitoring module include: acoustic Doppler current profiler, radar current meter, hydrological buoy, rainfall sensor, and K-band planar radar flow meter.

5. The dynamic monitoring system for dredging in waterway improvement projects according to claim 1, characterized in that, The monitoring of ecological and environmental changes includes the monitoring of water quality and bottom sediment; The waterway monitoring module uses a multi-parameter water quality sensor to monitor water quality, including pH, dissolved oxygen, turbidity, conductivity, temperature, COD, and ammonia nitrogen. Suspended solids or heavy metals are rapidly sampled from river water or silt using a monitoring submersible, and then tested in a laboratory.

6. The dynamic monitoring system for dredging in waterway improvement projects according to claim 1, characterized in that, The dredging operation monitoring module includes: GNSS or Beidou positioning components, dredging robot positioning components, high-frequency imaging sonar, draft sensor, AIS and ship radar.

7. The dynamic monitoring system for dredging in waterway improvement projects according to claim 1, characterized in that, The intelligent monitoring module includes an edge computing node, a cloud server, a data storage device, and a display; the edge computing node connects the sensor and the cloud server, filters the data collected by the sensor, and uploads it to the cloud server; An AI inference engine and a digital twin engine are deployed on the cloud server. The digital twin engine integrates multi-source data in real time, calculates and generates a digital image of the river, and completes 3D scene rendering on the display for automated real-time panoramic monitoring. The AI ​​inference engine and the digital twin engine work together to drive intelligent analysis and risk warning, realizing a closed loop from data to decision-making, enabling intelligent control, automatic analysis, and automatic decision-making.

8. A method for dynamic monitoring of dredging in waterway improvement projects according to any one of claims 1 to 7, characterized in that, The dynamic monitoring method includes the following steps: S1. Based on the existing monitoring sensors and equipment in the waterway, reasonably deploy the remaining sensors and equipment that need to be fixedly installed in the waterway monitoring module to monitor the mud surface elevation, silt thickness, siltation rate, minimum navigable depth, flow velocity and direction, sediment content or sediment inflow, silt density or rheology, and water quality ecology for a long period of time. S2. For sensors and equipment that require mobile monitoring in the waterway monitoring module, equip them with corresponding monitoring vessels and submersibles to carry out waterway inspection operations, collect data on full-section water depth topography, silt thickness distribution, cross-sectional flow velocity distribution, and water quality cross-section distribution, and sample river water and bottom sediment for testing and analysis; edge computing nodes receive data, perform data cleaning and screening locally, and upload core data to the cloud server; S3, the digital twin engine on the cloud server integrates a 3D model of the riverbed, presenting the siltation distribution and cross-sectional shape in real time. Sensor data drives the twin to update in real time, providing panoramic mapping; dynamic simulation of the effects of different dredging schemes, allowing selection of a scheme based on actual needs. S4, the AI ​​inference engine intelligently identifies the siltation range and predicts the dredging priority. The intelligent monitoring module uses big data and artificial intelligence technology to analyze and predict monitoring data. Through data models and algorithms, it identifies the risk of waterway siltation, predicts the dredging cycle and the priority of dredging work, and, combined with water level prediction, analyzes the siltation area at the corresponding water level and formulates the dredging tasks for each section. The S5 AI inference engine plans the construction path based on the dredging task, outputs decision reports or instructions, and after confirmation by management and construction personnel, the construction execution terminal controls the construction equipment to work. During the dredging process, the AI ​​inference engine automatically recommends the optimal parameters, optimizes them automatically, and remotely controls them based on real-time working conditions, silt disturbance and diffusion, and water quality ecology. Important decisions such as dredging plans or dredging schedules are made by using a digital twin engine to simulate multiple options, an AI inference engine to screen and compare the advantages and disadvantages of each option, and finally, the decision is made by the management personnel. S6. Based on the dredging task, dredging is carried out in segments with precision to reduce damage to the ecology. The dredging equipment used includes dredging boats and dredging robots. The dredging boats are the main equipment for targeted and efficient dredging of the core siltation points with high dredging priority. The dredging robots are used as an auxiliary equipment to supplement the dredging of areas missed by the dredging boats, thereby expanding the dredging range. At the same time, they can carry out fine dredging of areas or thin mud layers that are inconvenient for dredging boats to reach. S7. Dredging data is transmitted back to the cloud server for centralized analysis, driving dynamic adjustments to the dredging strategy in real time. The cloud server also undertakes the task of training the AI ​​model. After the AI ​​model is trained, it is compressed and distributed to the edge computing node of the dredging robot to complete the model iteration and upgrade. S8. In the area where dredging has been completed, the monitoring vessel or monitoring submersible will conduct an inspection. The multi-source data collected by the waterway monitoring module will be integrated, and the dredging effect will be jointly verified by the AI ​​inference engine and the digital twin engine. The results will be determined to determine whether the contaminated layer is clean and whether any soil has been mistakenly excavated. The 3D models before and after dredging will be automatically compared to accurately calculate the dredging volume and water depth recovery. Meanwhile, the AI ​​inference engine learns the siltation pattern based on historical siltation data and hydrodynamic data, predicts the future siltation rate, automatically generates suggestions for the next dredging time, and performs predictive proactive dredging.

9. The method for dynamic monitoring of dredging in waterway improvement projects according to claim 8, characterized in that, In S6, the dredging vessel and the dredging robot adopt a dual-platform collaborative mode. Different sensors and operating equipment are configured on each dredging vessel or dredging robot as needed and according to the budget to achieve joint operation. The dredging vessel is equipped with a dredging robot positioning component to monitor the robot's position in real time and schedule the operation path of multiple machines. The dredging vessel is equipped with multibeam sonar to perform large-area scanning, supplemented by side-scan sonar to check underwater debris and detect siltation patterns, providing precise underwater sensing support for dredging operations; The dredging robot is equipped with high-frequency imaging sonar and uses real-time reasoning based on edge computing. It extracts acoustic image features through convolutional neural networks, identifies silt textures and obstacle shapes, and generates a three-dimensional real-scene topographic map. The system can autonomously perceive, plan paths, and precisely control depth to achieve precise dredging that only removes the contaminated layer and avoids obstacles.

10. A method for dynamic monitoring of dredging in waterway improvement projects according to claim 8, characterized in that, The monitoring submersible or dredging robot is also equipped with an attitude detection module, which includes an IMU inertial measurement unit, an electronic compass, a depth / pressure sensor and a 9-axis attitude sensor. Combined with side-scan sonar and multi-beam sonar, it provides external reference and helps to correct accumulated errors. The monitoring submersible or dredging robot is connected to an unmanned vessel, which is equipped with an electric winch. A recovery rope is wound around the electric winch, and one end of the recovery rope is connected to the upper side of the monitoring submersible or dredging robot to facilitate the correction of the attitude of the monitoring submersible or dredging robot, or to quickly recover the monitoring submersible or dredging robot.

Images