Channel dredging intelligent monitoring system based on Beidou positioning

Abstract

The invention discloses a channel dredging intelligent monitoring system based on Beidou positioning. The system comprises a data fusion module, a dynamic grid module, a coverage analysis module, a siltation prediction module and a monitoring generation module. The data fusion module collects channel data through Beidou positioning and a multi-source sensor, and generates a real-time three-dimensional terrain model; the dynamic grid module divides dynamic grids according to the three-dimensional terrain model and dredging equipment parameters; the coverage analysis module calculates the coverage degree with the dynamic grid by using the operation track of the dredging equipment, and performs identification to obtain missing area data; the siltation prediction module predicts siltation distribution data by using a neural network model based on layered flow velocity data in the channel data and a historical siltation database; and the monitoring generation module fuses the three-dimensional terrain model, the missing area data and the deposition distribution data to generate a dredging monitoring report. Intelligent monitoring of channel dredging operation is achieved, and the monitoring efficiency and accuracy are improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of water transport engineering, and in particular relates to an intelligent monitoring system for waterway dredging based on Beidou positioning. Background Art

[0002] With the rapid development of inland waterway shipping and marine engineering technologies, the need for intelligent monitoring of waterway dredging operations is becoming increasingly urgent. In recent years, the high-precision positioning capabilities of the Beidou satellite navigation system have provided new technical means for waterway mapping, and multi-source sensor fusion technology has been widely applied in terrain modeling. Traditional waterway dredging monitoring primarily relies on a combination of GPS single-point positioning and manual bathymetry. Operational area coverage analysis is performed using a fixed grid, and empirical models are used to predict sediment accumulation distribution. Traditionally, dredging operation monitoring typically employs a static gridding approach, discretizing the waterway area based on a preset fixed-size grid. Equipment trajectory data is recorded at discrete sampling points, and a simple geometric overlap algorithm is used to determine grid coverage. Accumulation predictions primarily rely on historical average flow velocity data and linear regression models, with risk factor weights determined through manual experience. Monitoring report generation suffers from data silos across multiple systems, requiring manual overlay and analysis of 3D terrain data, equipment trajectory data, and prediction results. However, existing monitoring methods have significant flaws: static meshing struggles to adapt to changes in dredging equipment operating parameters and terrain undulations, resulting in redundant meshes in flat areas and insufficient resolution in complex areas; single GPS positioning data exhibits signal drift in bridge-blocked waters, resulting in temporal and spatial reference deviations from bathymetric sensor data; trajectory coverage analysis algorithms fail to consider the dynamic effects of equipment operating speed and sweep width, resulting in inaccurate identification of missed areas; traditional sedimentation prediction models fail to effectively integrate stratified velocity field data with multi-dimensional environmental factors, resulting in large prediction errors; and the lack of a dynamic optimization mechanism for dredging paths based on real-time data leads to duplicated operations and wasted energy. These issues severely restrict the monitoring efficiency and quality control level of waterway dredging operations. Summary of the Invention

[0003] Based on this, it is necessary to provide a waterway dredging intelligent monitoring system based on Beidou positioning that can solve the above problems.

[0004] In the first aspect, the present application provides a waterway dredging intelligent monitoring system based on Beidou positioning, comprising:

[0005] The data fusion module is used to collect channel data through Beidou positioning and multi-source sensors to generate a real-time three-dimensional terrain model;

[0006] Dynamic grid module, used to divide dynamic grid according to 3D terrain model and dredging equipment parameters;

[0007] Coverage analysis module, which uses the operation trajectory of dredging equipment to calculate the coverage with the dynamic grid and identify the missing area data;

[0008] The sedimentation prediction module is used to predict the sedimentation distribution data using a neural network model based on the layered velocity data in the waterway data and the historical sedimentation database;

[0009] The monitoring generation module is used to integrate the three-dimensional terrain model, missing area data and siltation distribution data to generate dredging monitoring reports.

[0010] In one embodiment, the coverage analysis module is further configured to:

[0011] Determine whether the density of sensor collection points in the dynamic grid meets the preset threshold requirements;

[0012] For dynamic grids that meet the accuracy requirements, the trajectory data of dredging equipment is used to calculate the coverage parameters of the trajectory data in the grid cells using the trajectory overlap analysis algorithm;

[0013] According to the coverage parameters, the regional difference algorithm is used to detect the uncovered dynamic grid and identify the missing area data.

[0014] In one embodiment, the coverage analysis module is further configured to:

[0015] The trajectory overlap analysis algorithm is constructed using the following formula:

[0016]

[0017] Among them, C ij represents the coverage index of grid (i, j), w(t) represents the time weight function, ΔL i ΔL j Represents the dynamic grid area, and 0.5L min ≤ΔL i , ΔL j ≤2L max , L min and L max They represent the minimum and maximum resolution of the sensor accuracy, P(t) represents the real-time position coordinates of the dredging equipment, and c ij represents the coordinates of the center point of the dynamic grid, σ represents the coverage influence radius adjustment coefficient, and the integration interval of this formula is [t1, t2], which is determined by the time range of the analysis period.

[0018] In one embodiment, the system further includes a dredging operation optimization module for:

[0019] According to the missing area data and siltation distribution data, a multi-objective genetic algorithm is used to optimize the path parameters of the dredging equipment and generate an initial path set.

[0020] A dynamic fitness function including coverage weight, sedimentation risk coefficient and energy consumption factor is constructed, and the initial path set is iteratively screened and cross-mutated to obtain the optimized path;

[0021] Dynamic conflict detection is performed on the optimized path based on the real-time three-dimensional terrain model, and the optimal operation path that meets the coverage threshold and energy consumption constraints is output.

[0022] In one embodiment, the data fusion module is further configured to:

[0023] Based on the error characteristics of Beidou positioning and multi-source sensors, a dynamic data screening mechanism is established to calibrate the temporal and spatial reference deviations of different sensors in real time to generate calibrated multi-source data.

[0024] Based on the calibrated multi-source data, a time-attenuation weighted multi-source fusion strategy is adopted to dynamically adjust the fusion weights according to the sensor accuracy level and the freshness of the Beidou positioning data to generate a weighted fusion dataset.

[0025] Based on the weighted fusion dataset, a three-dimensional terrain model with confidence classification is constructed.

[0026] In one embodiment, the dynamic grid module is further configured to:

[0027] Generate the initial dynamic grid distribution based on the slope distribution data of the 3D terrain model and the operating capacity parameters of the dredging equipment;

[0028] According to the system computing resource constraints, the accuracy and quantity of the initial dynamic grid distribution are balanced and optimized to obtain the intermediate optimized grid distribution;

[0029] By analyzing the terrain continuity characteristics between adjacent grids, the grid unit merging operation is performed on the flat area to generate the merged grid distribution;

[0030] The optimized dynamic grid is generated based on the intermediate optimized grid distribution and the merged grid distribution.

[0031] In one embodiment, the system further includes a visualization module for:

[0032] Based on the 3D terrain model, the locations of the missed areas and silt distribution are overlaid and rendered to generate a 3D map of the waterway;

[0033] Based on the coverage index of dynamic grids, a grid coloring algorithm is used to generate a dredging coverage heat map;

[0034] By utilizing the preset interactive mechanism, combined with the three-dimensional map of the waterway and the dredging coverage heat map, a multi-dimensional visual monitoring interface with temporal and spatial correlation is generated.

[0035] Secondly, this application also provides a method for intelligent monitoring of waterway dredging based on Beidou positioning, including:

[0036] Using Beidou positioning and multi-source sensors to collect channel data, a real-time three-dimensional terrain model is generated;

[0037] Divide dynamic grids according to 3D terrain model and dredging equipment parameters;

[0038] Using the dredging equipment's operating trajectory, the coverage with the dynamic grid is calculated to identify the missing area data.

[0039] Based on the stratified velocity data in the waterway data and the historical sedimentation database, the siltation distribution data is predicted using a neural network model;

[0040] The 3D terrain model, missing area data and siltation distribution data are integrated to generate dredging monitoring reports.

[0041] On the third aspect, the present application also provides a computer device including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, it implements the steps of the above-mentioned function of the waterway dredging intelligent monitoring system based on Beidou positioning.

[0042] Fourthly, the present application also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of implementing the functions of the above-mentioned Beidou positioning-based waterway dredging intelligent monitoring system are implemented.

[0043] The above-mentioned intelligent monitoring system, method, computer equipment and storage medium for channel dredging based on Beidou positioning, the data fusion module integrates Beidou positioning and multi-source sensor data, overcomes the drift problem of traditional positioning signals, and constructs a high-precision real-time three-dimensional terrain model; the dynamic grid module adaptively divides the grid according to terrain characteristics and equipment parameters, and solves the resolution imbalance defect of static grids in complex terrain; the coverage analysis module accurately identifies missed areas based on the spatial coupling relationship between equipment trajectory and dynamic grid, thereby improving the recognition accuracy; the siltation prediction module uses a neural network to fuse layered velocity field and historical siltation data, breaking through the limitations of the linear regression model and reducing prediction errors; the monitoring generation module automatically generates structured reports through the fusion of multi-dimensional data, realizes closed-loop control of operation quality, and significantly improves the real-time and scientific decision-making of channel dredging monitoring. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the following briefly introduces the drawings required for use in the embodiments or related technical descriptions. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0045] Figure 1 This is a structural diagram of the waterway dredging intelligent monitoring system based on Beidou positioning of the present invention;

[0046] Figure 2 This is a flow chart of the intelligent monitoring method for waterway dredging based on Beidou positioning of the present invention. DETAILED DESCRIPTION

[0047] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0048] This application presents an intelligent waterway dredging monitoring system based on Beidou positioning, consisting of Beidou positioning equipment, multi-source sensors, a computing terminal, and a server, connected via a wired or wireless network. During river dredging and port dredging, Beidou and sensors collect waterway data in real time. The computing terminal dynamically optimizes grid divisions and identifies missed areas. The server predicts siltation risks based on real-time data, optimizes the path, and feeds it back to the dredging vessel control system, achieving closed-loop control of dredging coverage and risk avoidance.

[0049] In one embodiment, Figure 1 As shown, a waterway dredging intelligent monitoring system based on Beidou positioning is provided. This embodiment uses the system deployed on a computing terminal as an example for illustration. It is understandable that the method can also be deployed on a server, or on an architecture including a computing terminal and a server, and implemented through the interaction between the computing terminal and the server. In this embodiment, the system includes the following modules:

[0050] The data fusion module 101 is used to collect channel data through Beidou positioning and multi-source sensors to generate a real-time three-dimensional terrain model.

[0051] The BeiDou positioning system collects spatial location information of the waterway. Multi-source sensors, such as depth sounders, side-scan sonars, and current meters, collect multi-dimensional information such as waterway depth, riverbed topography, and water velocity. After data processing such as data screening, spatiotemporal benchmark calibration, and weighted fusion, a three-dimensional terrain model can be constructed using triangulated or regular grid modeling methods to present the three-dimensional topography of the waterway, such as water depth and riverbed undulations, and the distribution of underwater obstacles. Using BeiDou positioning data and waterway data collected by multi-source sensors, through calibration, fusion processing, and three-dimensional modeling, the real-time three-dimensional terrain model generated can provide basic data support for subsequent modules such as grid division and coverage analysis, providing a decision-making basis for the planning, execution, and supervision of dredging operations.

[0052] The dynamic grid module 102 is used to divide the dynamic grid according to the three-dimensional terrain model and the dredging equipment parameters.

[0053] Among them, the grid can be dynamically divided according to the terrain information provided by the three-dimensional terrain model, taking into account the parameters of the dredging equipment such as the three-dimensional parameters of the equipment, excavation depth, excavation range and applicable environment. The grid can be divided only for the area of ​​the operating capacity of the dredging equipment, and twice the maximum operating width of the dredging equipment, such as 10m×10m, is used as the reference grid size, and dynamically adjusted according to the terrain information of the three-dimensional terrain model. For example, for areas with a slope ≥15°, the grid resolution is increased to 1 / 2 of the maximum operating width of the dredging equipment, such as 2.5m×2.5m. By dynamically dividing the grid in this way, the number of redundant grids can be reduced, and the accuracy of capturing the features of complex terrain can be improved, laying a spatial benchmark for subsequent coverage analysis and siltation prediction.

[0054] The coverage analysis module 103 is used to calculate the coverage with the dynamic grid using the operation trajectory of the dredging equipment, and identify the missing area data.

[0055] Dynamic grids are constructed based on a three-dimensional terrain model and dredging equipment parameters, taking into account terrain characteristics and equipment operating capabilities, providing a reasonable spatial unit for coverage analysis. Dredging equipment trajectory data is acquired in real time by positioning systems such as Beidou, recording the equipment's movement path during operation and forming the basis for the coverage analysis module. When calculating the coverage of a dynamic grid, multiple factors can be considered, including a time weighting function, the area of the dynamic grid, and the minimum and maximum resolution of sensor accuracy. For example, the time weighting function assigns different weights to trajectory data at different times, reflecting the varying importance of equipment operations to grid coverage at different times. This comprehensive consideration allows for a more accurate measurement of the coverage of each dynamic grid by the equipment trajectory, resulting in a grid coverage index, a quantitative indicator of coverage status that provides data support for subsequent analysis. Based on the calculated coverage parameters, uncovered dynamic grids are used to identify areas with low or no coverage, generating data on missed areas. This data allows operators to adjust the dredging equipment's operating path in a timely manner and re-dredge missed areas to ensure that the waterway dredging meets the expected standards.

[0056] The sedimentation prediction module 104 is used to predict sedimentation distribution data using a neural network model based on the layered velocity data in the waterway data and the historical sedimentation database.

[0057] Among them, water flow velocity has a direct impact on the transport and accumulation of waterway sediment, and the flow velocity varies in different water layers. The layered flow velocity data can reflect the flow characteristics of water in the vertical direction, which helps to analyze the movement patterns of sediment in different water layers. The historical sedimentation database contains records of the sedimentation of waterways in different time periods, covering information such as the location, thickness, and area of sedimentation, reflecting the historical evolution of waterway sedimentation, and understanding the seasonal changes, periodic patterns, and associations with different hydrological conditions of waterway sedimentation. Through data learning, the neural network model can capture the comprehensive impact of various factors such as flow velocity, direction, water depth, sediment content, and topography on sedimentation, generate predicted sedimentation distribution data, reflect changes in waterway sedimentation, and provide support for real-time management of waterways.

[0058] The monitoring generation module 105 is used to fuse the three-dimensional terrain model, the missing area data and the siltation distribution data to generate a dredging monitoring report.

[0059] A 3D terrain model can be used as a framework to derive a basic spatial coordinate system for the waterway. Data on missed areas can then be overlaid on this foundation. This allows the terrain model to display areas that were missed during dredging operations. This, combined with siltation distribution data, identifies areas at risk of future siltation, creating a comprehensive and integrated information set encompassing both the current status of the waterway, operational issues, and future trends. The resulting dredging monitoring report can include both textual descriptions and graphical presentations. The textual descriptions describe the overall topography of the waterway, the specific locations and causes of missed areas during dredging operations, and areas at risk of future siltation and their potential impact on the waterway's navigability. Graphical presentations can include graphical formats such as 3D maps, bar charts, and line graphs. For example, these can color-code missed areas and areas at risk of siltation, and use bar charts to compare predicted siltation thicknesses across different regions. The resulting dredging monitoring report can provide important decision-making support for waterway management and dredging teams. Rationally arrange subsequent dredging plans, determine key dredging areas and time nodes, optimize resource allocation, improve waterway maintenance efficiency, and adjust operation plans in a timely manner based on the missed areas identified to ensure the comprehensiveness and accuracy of dredging operations and ensure the safety and smooth flow of waterways.

[0060] The above-mentioned intelligent monitoring system for waterway dredging based on Beidou positioning constructs a real-time three-dimensional terrain model by integrating Beidou positioning and multi-source sensor data, overcoming the problems of traditional GPS positioning signal drift and multi-source data spatiotemporal benchmark deviation; adopts a dynamic grid division mechanism combined with dredging equipment parameters and terrain characteristics to achieve adaptive adjustment of grid resolution, solving the coverage redundancy and accuracy imbalance defects of static grids in complex terrain; quantifies the spatial coupling relationship between equipment trajectory and dynamic grid, identifies missed areas through coverage calculation, and improves the accuracy of operation coverage analysis; uses a neural network model to fuse layered flow velocity data and historical siltation characteristics, breaking through the limitations of traditional linear regression models in capturing the correlation of multi-dimensional environmental factors and reducing siltation prediction errors; automatically generates structured monitoring reports through multi-dimensional data fusion, realizing closed-loop management of dredging operation quality assessment, missed area re-excavation and siltation risk warning, effectively improving the real-time and scientific decision-making of waterway dredging monitoring, and solving the problems of frequent manual intervention, serious data siltation and insufficient dynamic optimization capabilities in traditional methods.

[0061] In one embodiment, the coverage analysis module 103 is further configured to:

[0062] Determine whether the density of sensor collection points in the dynamic grid meets the preset threshold requirements;

[0063] For dynamic grids that meet the accuracy requirements, the trajectory data of dredging equipment is used to calculate the coverage parameters of the trajectory data in the grid cells using the trajectory overlap analysis algorithm;

[0064] According to the coverage parameters, the regional difference algorithm is used to detect the uncovered dynamic grid and identify the missing area data.

[0065] Specifically, sensor point density is a key indicator of data accuracy. Before conducting coverage analysis, the sensor point density within a dynamic grid is determined to have met a preset threshold. This threshold is set based on the accuracy requirements for waterway monitoring, sensor performance, and actual operational conditions. If the point density meets the threshold, the dynamic grid's data accuracy is sufficient, providing a reliable foundation for subsequent coverage analysis. If it does not, it indicates that the data may be missing or inaccurate. Such grids, lacking data reliability, will be excluded from subsequent coverage parameter calculations to avoid erroneous analysis results due to data quality issues. For dynamic grids that meet accuracy requirements, coverage parameters are calculated using the dredging equipment trajectory data using a trajectory overlap analysis algorithm. This algorithm considers multiple factors, such as the equipment's operating speed, which affects its dwell time and coverage range within each grid cell; the sweep width, which determines the lateral area covered by the equipment in a single operation; and time, assigning different weights to trajectory data from different time periods to more accurately reflect the actual impact of equipment operations on grid coverage. This calculation yields coverage parameters, such as coverage area and number of coverages, that quantify the degree of coverage of each dynamic grid cell by the dredging equipment trajectory. A regional difference algorithm is used to detect uncovered dynamic grids. By comparing the coverage parameters of adjacent grids or different grids within the entire region, the regional difference algorithm identifies grids with significantly lower coverage than other areas. These areas, known as areas that are not effectively covered, may be due to improper dredging equipment path planning, equipment failure, and other reasons. These uncovered dynamic grids are marked, and missing area data is identified. The location, scope, and other information of these missing areas are recorded. Based on this missing area data, waterway management departments and operation teams can promptly adjust the dredging equipment's operation path and plan, supplement the dredging work in the missing areas, and ensure the comprehensiveness and effectiveness of waterway dredging.

[0066] In one embodiment, the coverage analysis module 103 is further configured to:

[0067] The trajectory overlap analysis algorithm is constructed using the following formula:

[0068]

[0069] Among them, C ij represents the coverage index of grid (i, j), w(t) represents the time weight function, ΔL i ΔL j Represents the dynamic grid area, and 0.5L min ≤ΔL i , ΔL j ≤2Lmax , L min and L max They represent the minimum and maximum resolution of the sensor accuracy, P(t) represents the real-time position coordinates of the dredging equipment, and c ij represents the coordinates of the center point of the dynamic grid, σ represents the coverage influence radius adjustment coefficient, and the integration interval of this formula is [t1, t2], which is determined by the time range of the analysis period.

[0070] For example, the coverage index C of the grid (i, j) is ij It is used to measure the degree to which the grid (i, j) is covered by the dredging equipment trajectory. The specific form and parameters of the time weight function w(t) can be determined according to the actual situation of the dredging operation and the influence of time on the coverage effect. For example, considering the operating efficiency and intensity of the dredging equipment in different time periods, through the analysis of historical operation data, a suitable time weight function form is fitted to reflect the difference in the importance of trajectory data at different time points to the grid coverage. Dynamic grid area ΔL i ΔL j It is determined based on factors such as terrain complexity and equipment operation capability. It is a basic parameter for calculating coverage index. The size of the grid area affects the calculation result of coverage. The minimum and maximum resolution of the sensor accuracy L min and L max Depends on the performance indicators of the sensors used, and is used to consider the impact of sensor accuracy on coverage calculations. During the calculation process, certain corrections are made to the coverage calculation results to ensure that the calculation results can more accurately reflect the actual coverage situation. For example, if the sensor accuracy is low, it may cause certain errors in the judgment of the coverage situation. By introducing these two parameters, this error can be compensated to a certain extent. The real-time position coordinates P(t) of the dredging equipment are obtained in real time by a positioning system such as the Beidou positioning system to determine the position of the equipment relative to the dynamic grid at each moment, determine whether the equipment has entered the grid and its movement trajectory within the grid, and affect the coverage calculation of the grid. The center point coordinates c of the dynamic grid ijUsed to compare with the real-time position coordinates of the dredging equipment to determine the relative positional relationship between the equipment and the grid. By calculating parameters such as the distance between the equipment position and the grid center point, the range and degree of the equipment's coverage of the grid can be determined. The coverage impact radius adjustment coefficient σ is set based on actual operating conditions and experience. Through testing and analysis of dredging operations under different working conditions, the appropriate adjustment coefficient value can be determined to adjust the range of influence of the dredging equipment on the grid coverage. Different dredging equipment may have different operating coverage ranges. By adjusting the value, the coverage calculation can be made more realistic. For example, for equipment with a larger coverage range, the σ value can be appropriately increased; for equipment with a smaller coverage range, the σ value can be reduced. The integration interval is determined by the time range of the analysis period [t1, t2], representing the time range considered when calculating the coverage index, ensuring that the calculation results reflect the grid coverage within a specific time period.

[0071] In one embodiment, the system further includes a dredging operation optimization module for:

[0072] According to the missing area data and siltation distribution data, a multi-objective genetic algorithm is used to optimize the path parameters of the dredging equipment and generate an initial path set.

[0073] A dynamic fitness function including coverage weight, sedimentation risk coefficient and energy consumption factor is constructed, and the initial path set is iteratively screened and cross-mutated to obtain the optimized path;

[0074] Dynamic conflict detection is performed on the optimized path based on the real-time three-dimensional terrain model, and the optimal operation path that meets the coverage threshold and energy consumption constraints is output.

[0075] Specifically, the multi-objective genetic algorithm simulates natural selection and genetic mechanisms, and simultaneously handles multiple optimization objectives, such as covering missed areas, avoiding areas with high siltation risk, etc. The path parameters of the dredging equipment are encoded into chromosomes to form an initial population, and each chromosome represents a possible dredging operation path. For example, a series of waypoints are randomly selected within the waterway range and connected into different paths as individuals of the initial population to generate an initial path set. A fitness function containing multiple objectives is constructed. In the waterway dredging scenario, the fitness function may include coverage weights, siltation risk coefficients, and energy consumption factors. For each chromosome in the population, that is, each initial path, its fitness value is calculated according to the fitness function. The higher the fitness value, the better the path performs in terms of comprehensively considering multiple objectives. A selection operation can be used to select a subset of individuals from the current population as parents. A crossover operation can then be performed on these selected parents to generate offspring individuals. For example, in waterway dredging path optimization, suppose a single-point crossover is used to randomly select an intersection point. The sections of the two parent paths after the intersection are swapped to generate two offspring paths. This can combine the advantages of different parent paths to produce a new, potentially more optimal path. Mutation operations can increase population diversity and prevent the algorithm from falling into local optima. This can be manifested as making small random adjustments to the coordinates of a particular waypoint. For example, the longitude or latitude of a waypoint can be randomly changed with a small probability to generate a new path. The selection, crossover, and mutation operations are repeated until a termination condition is met. Termination conditions can include reaching a maximum number of iterations or no significant improvement in fitness, resulting in the optimized path. After dynamic conflict detection, the path that meets a coverage threshold (i.e., a certain coverage requirement for missed areas) and an energy consumption constraint (i.e., the energy consumption of the equipment on the path does not exceed a preset upper limit) is selected as the optimal operating path, providing guidance for actual dredging equipment operations.

[0076] In one embodiment, the data fusion module 101 is further configured to:

[0077] Based on the error characteristics of Beidou positioning and multi-source sensors, a dynamic data screening mechanism is established to calibrate the temporal and spatial reference deviations of different sensors in real time to generate calibrated multi-source data.

[0078] Based on the calibrated multi-source data, a time-attenuation weighted multi-source fusion strategy is adopted to dynamically adjust the fusion weights according to the sensor accuracy level and the freshness of the Beidou positioning data to generate a weighted fusion dataset.

[0079] Based on the weighted fusion dataset, a three-dimensional terrain model with confidence classification is constructed.

[0080] For example, Beidou positioning and multi-source sensors are subject to various factors when collecting data, leading to errors. For example, Beidou positioning may experience signal drift in waters obscured by bridges, and different sensors also have varying measurement accuracy and time response. Based on these error characteristics, a dynamic data screening mechanism is established to monitor data quality in real time and remove anomalous data according to pre-set rules. For example, when data collected by a sensor exceeds a reasonable range or its trend does not conform to actual conditions, it is considered anomalous and screened out. Furthermore, a specific algorithm is used to perform spatiotemporal calibration to ensure temporal and spatial consistency of data collected by different sensors. A time-decrease weighted multi-source fusion strategy dynamically adjusts fusion weights based on the sensor accuracy level and the freshness of Beidou positioning data. This weighting is assigned to high-precision sensor data or fresh Beidou positioning data, and gradually decreases over time, emphasizing the timeliness of the data. This generates a weighted fusion dataset that leverages the strengths of different sensors and improves data accuracy and reliability. Each data point or region in the model is assigned a different confidence level based on factors such as the data source, accuracy, and weighting during the fusion process. For example, areas with high-precision sensor data and high weighting receive a higher confidence level, while areas with relatively poor data quality receive a lower confidence level. When using 3D terrain models for analysis and decision-making, users can assess data reliability based on confidence levels, improving the scientific nature of their decisions.

[0081] In one embodiment, the dynamic grid module 102 is further configured to:

[0082] Generate the initial dynamic grid distribution based on the slope distribution data of the 3D terrain model and the operating capacity parameters of the dredging equipment;

[0083] According to the system computing resource constraints, the accuracy and quantity of the initial dynamic grid distribution are balanced and optimized to obtain the intermediate optimized grid distribution;

[0084] By analyzing the terrain continuity characteristics between adjacent grids, the grid unit merging operation is performed on the flat area to generate the merged grid distribution;

[0085] The optimized dynamic grid is generated based on the intermediate optimized grid distribution and the merged grid distribution.

[0086] Specifically, the slope distribution data of the 3D terrain model reflects the undulations of the waterway topography. Areas with complex terrain and large slope variations may contain special terrain features such as reefs and gullies, requiring finer meshing for accurate monitoring. Dredging equipment operating capacity parameters include dredging depth, dredging range, and movement speed. If the equipment's dredging range is large, the corresponding mesh size should also be relatively large to ensure accurate coverage of the equipment's operating area. For example, a large trailing suction hopper dredger has a large operating range, so the mesh size can be appropriately increased based on its operating capacity. An initial dynamic mesh distribution is generated by comprehensively considering the slope distribution data and the dredging equipment's operating capacity parameters. Based on system computing resource constraints, the accuracy and number of the initial dynamic mesh distribution are optimized to balance monitoring accuracy. While maintaining a certain level of monitoring accuracy, the number of meshes can be appropriately reduced, or within the limits of computing resources, mesh accuracy can be moderately increased. This approach yields an intermediate optimized mesh distribution that meets computing resource constraints while ensuring a certain degree of monitoring accuracy. The terrain continuity between adjacent meshes reflects the gradualness of terrain changes. For areas with flat terrain and minimal changes, the terrain data differences between adjacent meshes are minimal. To reduce data processing and improve computational efficiency, grid cell merging can be performed on flat areas. For example, in shallow areas with relatively uniform siltation, multiple adjacent small grids can be merged into a single large grid. This reduces the number of grid cells and computational complexity without losing important topographic information, generating a merged grid distribution. By comprehensively considering the intermediate optimized grid distribution and the merged grid distribution, we can ensure monitoring accuracy within computational resource constraints, optimize the grid division in flat areas, and achieve efficient data processing and monitoring.

[0087] In one embodiment, the system further includes a visualization module for:

[0088] Based on the 3D terrain model, the locations of the missed areas and silt distribution are overlaid and rendered to generate a 3D map of the waterway;

[0089] Based on the coverage index of dynamic grids, a grid coloring algorithm is used to generate a dredging coverage heat map;

[0090] By utilizing the preset interactive mechanism, combined with the three-dimensional map of the waterway and the dredging coverage heat map, a multi-dimensional visual monitoring interface with temporal and spatial correlation is generated.

[0091] For example, the locations of missed areas and siltation distribution can be overlaid and rendered on a 3D terrain model. Missing and silted areas can be highlighted using different colors, textures, or marking methods. For example, missing areas can be marked in red, while areas with a higher risk of siltation can be marked in yellow, allowing users to quickly identify the location and scope of key areas of concern in the waterway in 3D space. A grid coloring algorithm can be used to assign different colors to each dynamic grid based on the coverage index. For example, areas with high coverage can be represented in green, areas with medium coverage in yellow, and areas with low coverage in red. This generates a dredging coverage heat map, which visually demonstrates the dredging coverage of the entire waterway area, allowing users to quickly identify areas with good and insufficient coverage, providing a clear basis for subsequent adjustments to dredging strategies. Pre-set interactive mechanisms allow users to interact with the visualization interface through mouse clicks, zooming, panning, and other operations. For example, clicking on an area on the 3D waterway map displays detailed information about that area, including depth, whether it is a missed area, and the level of siltation risk. Combining a 3D waterway map with a dredging coverage heat map creates a visual interface that takes time into account, showcasing changes in waterway status at different points in time. For example, users can switch timelines to view dredging progress, changes in missed areas, and the dynamic evolution of siltation distribution over time. This allows users to observe and analyze waterway dredging operations from multiple perspectives. They can view current spatial distribution information while also tracing historical data to understand the evolution of operations over time. This provides richer and more comprehensive information support for waterway management and decision-making, helping users identify problems promptly and make informed decisions.

[0092] The aforementioned Beidou-based intelligent dredging monitoring system utilizes Beidou positioning and multi-source sensor data to establish a dynamic data screening mechanism and calibrate spatiotemporal benchmark deviations. A time-degradation weighted fusion strategy is then used to construct a three-dimensional terrain model with confidence levels. This overcomes the issues of traditional positioning signal drift and multi-source data asynchrony, providing high-precision basic data. The dynamic meshing module creates a dynamic mesh based on the slope distribution data of the 3D terrain model and the operational capacity parameters of the dredging equipment. This optimization takes into account both computing resources and terrain continuity, addressing the resolution imbalance of traditional static meshing in complex terrain and making the meshing more tailored to actual operational needs. The coverage analysis module determines the density of sensor data points within the dynamic mesh and, using trajectory overlap analysis and regional difference algorithms, accurately calculates the coverage of dredging equipment trajectories with the dynamic mesh and identifies missed areas, improving the accuracy of missed area identification compared to traditional methods. The sedimentation prediction module uses a neural network model to predict sedimentation distribution data based on layered flow velocity data and a historical sedimentation database, overcoming the limitations of traditional linear regression models and reducing prediction errors. The monitoring generation module integrates 3D terrain models, data on omitted areas, and silt distribution data to generate dredging monitoring reports, achieving closed-loop control of operation quality. Furthermore, the dredging operation optimization module uses a multi-objective genetic algorithm to optimize dredging equipment path parameters based on omitted area and silt distribution data. It constructs a dynamic fitness function for iterative screening and cross-mutation, and performs dynamic conflict detection based on a real-time 3D terrain model. It outputs the optimal operation path that meets coverage thresholds and energy consumption constraints, thereby improving operation efficiency and reducing energy consumption and duplication of work. The visualization module generates a 3D map of the waterway by overlaying and rendering omitted areas and silt distribution locations on the 3D terrain model. It uses a dynamic grid coverage index to generate a dredging coverage heat map, and combines a preset interactive mechanism to generate a multi-dimensional visualization monitoring interface with temporal and spatial correlations. This provides users with intuitive and comprehensive monitoring information, enabling them to promptly grasp the operation status and make scientific decisions. This addresses the problems of traditional monitoring methods and is of great significance to the development of inland waterway shipping and marine engineering technology.

[0093] It should be understood that, although the various steps in the flowcharts involved in the various embodiments described above are displayed in sequence according to the instructions of the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and these steps can be executed in other orders. Moreover, at least a portion of the steps in the flowcharts involved in the various embodiments described above can include multiple steps or multiple stages, and these steps or stages are not necessarily executed and completed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a portion of steps or stages in other steps.

[0094] Based on the same inventive concept, the embodiments of the present application also provide a method for intelligent monitoring of waterway dredging based on Beidou positioning, which is used to implement the aforementioned intelligent monitoring system for waterway dredging based on Beidou positioning. The implementation solution provided by this device is similar to the implementation solution described in the aforementioned method. Therefore, the specific limitations of one or more embodiments of the method for intelligent monitoring of waterway dredging based on Beidou positioning provided below can be found in the above-mentioned limitations of the intelligent monitoring system for waterway dredging based on Beidou positioning, and will not be repeated here.

[0095] In an exemplary embodiment, Figure 2 As shown, a method for intelligent monitoring of waterway dredging based on Beidou positioning is provided, including:

[0096] S11, collects channel data through Beidou positioning and multi-source sensors to generate a real-time 3D terrain model;

[0097] S12, dividing the dynamic grid according to the three-dimensional terrain model and dredging equipment parameters;

[0098] S13, using the operation trajectory of the dredging equipment, calculating the coverage with the dynamic grid, and identifying the missing area data;

[0099] S14, based on the stratified velocity data in the waterway data and the historical siltation database, the siltation distribution data is predicted using a neural network model;

[0100] S15, integrating the three-dimensional terrain model, the missing area data and the siltation distribution data to generate a dredging monitoring report.

[0101] In one embodiment, the operation trajectory of the dredging equipment is used to calculate the coverage with the dynamic grid and identify the missing area data, including:

[0102] S21, determining whether the density of sensor collection points in the dynamic grid meets a preset threshold requirement;

[0103] S22, for the dynamic grid that meets the accuracy requirements, based on the trajectory data of the dredging equipment, a trajectory overlap analysis algorithm is used to calculate the coverage parameters of the trajectory data in the grid unit;

[0104] S23, based on the coverage parameters, using a regional difference algorithm to detect uncovered dynamic grids, and identifying missing area data.

[0105] In one embodiment, the method further comprises:

[0106] S31, construct the trajectory overlap analysis algorithm using the following formula:

[0107]

[0108] Among them, C ij represents the coverage index of grid (i, j), w(t) represents the time weight function, ΔL i ΔL j Represents the dynamic grid area, and 0.5L min ≤ΔL i , ΔL j ≤2L max , L min and L max They represent the minimum and maximum resolution of the sensor accuracy, P(t) represents the real-time position coordinates of the dredging equipment, and c ij represents the coordinates of the center point of the dynamic grid, σ represents the coverage influence radius adjustment coefficient, and the integration interval of this formula is [t1, t2], which is determined by the time range of the analysis period.

[0109] In one embodiment, the method further comprises:

[0110] S41, based on the missing area data and the siltation distribution data, a multi-objective genetic algorithm is used to optimize the path parameters of the dredging equipment to generate an initial path set;

[0111] S42, constructing a dynamic fitness function including coverage weight, sedimentation risk coefficient and energy consumption factor, and performing iterative screening and cross-mutation operations on the initial path set to obtain the optimized path;

[0112] S43, performing dynamic conflict detection on the optimized path based on the real-time three-dimensional terrain model, and outputting the optimal operation path that meets the coverage threshold and energy consumption constraints.

[0113] In one embodiment, channel data is collected through Beidou positioning and multi-source sensors to generate a real-time three-dimensional terrain model, including:

[0114] S51, based on the BeiDou positioning and multi-source sensor error characteristics, establishes a dynamic data screening mechanism and calibrates the spatiotemporal reference deviations of different sensors in real time to generate calibrated multi-source data;

[0115] S52, based on the calibrated multi-source data, adopts a time-attenuation weighted multi-source fusion strategy, dynamically adjusts the fusion weight according to the sensor accuracy level and the freshness of the Beidou positioning data, and generates a weighted fusion data set;

[0116] S53, constructs a three-dimensional terrain model with confidence classification based on the weighted fusion dataset.

[0117] In one embodiment, dividing the dynamic grid according to the three-dimensional terrain model and the dredging equipment parameters further comprises:

[0118] S61, generating an initial dynamic grid distribution based on the slope distribution data of the three-dimensional terrain model and the operating capacity parameters of the dredging equipment;

[0119] S62, based on the system computing resource constraints, balance optimization of the accuracy and quantity of the initial dynamic grid distribution to obtain an intermediate optimized grid distribution;

[0120] S63, by analyzing the terrain continuity characteristics between adjacent grids, performing a grid unit merging operation on the flat area to generate a merged grid distribution;

[0121] S64: Generate an optimized dynamic grid based on the intermediate optimized grid distribution and the merged grid distribution.

[0122] In one embodiment, the method further comprises:

[0123] S71, based on the three-dimensional terrain model, overlaying and rendering the positions of the missed areas and the siltation distribution to generate a three-dimensional map of the waterway;

[0124] S72, based on the dynamic grid coverage index, a grid coloring algorithm is used to generate a dredging coverage heat map;

[0125] S73, using the preset interactive mechanism, combined with the three-dimensional map of the waterway and the dredging coverage heat map, generates a multi-dimensional visual monitoring interface with temporal and spatial correlation.

[0126] In one embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of the waterway dredging intelligent monitoring system based on Beidou positioning as described above are implemented.

[0127] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0128] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to the partial description of the method embodiments. The device embodiments described above are merely illustrative, wherein the components described as separate parts may or may not be physically separated, and the parts displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the disclosed solution. A person of ordinary skill in the art can understand and implement it without expending creative work.

[0129] The above-described embodiments merely represent several implementation methods of the embodiments of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the patent application. It should be noted that a person skilled in the art may make various modifications and improvements without departing from the concept of the embodiments of the present application, and these modifications and improvements fall within the scope of protection of the embodiments of the present application.

Claims

1. A waterway dredging intelligent monitoring system based on Beidou positioning, characterized in that: The system comprises: The data fusion module is used to collect channel data through Beidou positioning and multi-source sensors to generate a real-time three-dimensional terrain model; A dynamic grid module, used for dividing a dynamic grid according to the three-dimensional terrain model and dredging equipment parameters; A coverage analysis module, configured to calculate the coverage of the dynamic grid using the operation trajectory of the dredging equipment, and identify and obtain missing area data; A siltation prediction module is used to predict siltation distribution data using a neural network model based on the layered flow velocity data in the waterway data and a historical siltation database; The monitoring generation module is used to fuse the three-dimensional terrain model, the missing area data and the siltation distribution data to generate a dredging monitoring report.

2. The system according to claim 1, wherein: The coverage analysis module is further configured to: Determining whether the density of sensor acquisition points in the dynamic grid meets a preset threshold requirement; For a dynamic grid that meets the accuracy requirements, based on the trajectory data of the dredging equipment, a trajectory overlap analysis algorithm is used to calculate the coverage parameters of the trajectory data in the grid unit; According to the coverage parameters, a regional difference algorithm is used to detect uncovered dynamic grids and identify missing area data.

3. The system according to claim 2, characterized in that The coverage analysis module is further configured to: The trajectory overlap analysis algorithm is constructed using the following formula: Among them, C ij represents the coverage index of grid (i, j), w(t) represents the time weight function, ΔL i ΔL j Represents the dynamic grid area, and 0.5L min ≤ΔL i , ΔL j ≤2L max , L min and L max They represent the minimum and maximum resolution of the sensor accuracy, P(t) represents the real-time position coordinates of the dredging equipment, and c ij represents the coordinates of the center point of the dynamic grid, σ represents the coverage influence radius adjustment coefficient, and the integration interval of this formula is [t1, t2], which is determined by the time range of the analysis period.

4. The system according to claim 1, wherein: The system also includes a dredging operation optimization module for: According to the omitted area data and the siltation distribution data, a multi-objective genetic algorithm is used to optimize the path parameters of the dredging equipment to generate an initial path set; A dynamic fitness function including coverage weight, sedimentation risk coefficient and energy consumption factor is constructed, and the initial path set is iteratively screened and cross-mutated to obtain the optimized path; Dynamic conflict detection is performed on the optimized path based on the real-time three-dimensional terrain model, and an optimal operation path that meets a coverage threshold and energy consumption constraints is output.

5. The system according to claim 4, wherein: The data fusion module is also used for: Based on the error characteristics of Beidou positioning and multi-source sensors, a dynamic data screening mechanism is established to calibrate the temporal and spatial reference deviations of different sensors in real time to generate calibrated multi-source data. Based on the calibrated multi-source data, a time-attenuation weighted multi-source fusion strategy is adopted to dynamically adjust the fusion weight according to the sensor accuracy level and the freshness of the Beidou positioning data to generate a weighted fusion data set; Based on the weighted fusion data set, a three-dimensional terrain model with confidence levels is constructed.

6. The system according to claim 1, wherein: The dynamic grid module is also used to: generating an initial dynamic grid distribution based on the slope distribution data of the three-dimensional terrain model and the operating capacity parameters of the dredging equipment; According to the system computing resource constraints, the accuracy and quantity of the initial dynamic grid distribution are balanced and optimized to obtain an intermediate optimized grid distribution; By analyzing the terrain continuity characteristics between adjacent grids, the grid unit merging operation is performed on the flat area to generate the merged grid distribution; An optimized dynamic grid is generated according to the intermediate optimized grid distribution and the merged grid distribution.

7. The system according to claim 1, wherein: The system further includes a visualization module for: Based on the three-dimensional terrain model, the positions of the missing areas and the silt distribution are superimposed and rendered to generate a three-dimensional map of the waterway; Based on the coverage index of the dynamic grid, a grid coloring algorithm is used to generate a dredging coverage heat map; By utilizing the preset interactive mechanism and combining the three-dimensional map of the waterway and the dredging coverage heat map, a multi-dimensional visual monitoring interface with temporal and spatial correlation is generated.

8. A method for intelligent monitoring of waterway dredging based on Beidou positioning, characterized in that: The method comprises: Using Beidou positioning and multi-source sensors to collect channel data, a real-time three-dimensional terrain model is generated; Dividing a dynamic grid according to the three-dimensional terrain model and dredging equipment parameters; Utilizing the operation trajectory of the dredging equipment, calculating the coverage with the dynamic grid, and identifying and obtaining the missing area data; Based on the stratified flow velocity data in the waterway data and the historical siltation database, a neural network model is used to predict and obtain siltation distribution data; The three-dimensional terrain model, the missing area data and the siltation distribution data are integrated to generate a dredging monitoring report.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the system function steps according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the system function according to any one of claims 1 to 7 are realized.

Images