A method, system and equipment for monitoring underwater siltation at a dock based on a BIM model

Abstract

The present invention provides a method, system and equipment for monitoring underwater siltation at a dock based on a BIM model, which relates to the field of marine engineering and intelligent monitoring technology. The method comprises the following steps: demarcating an underwater siltation monitoring area based on the dock BIM model coordinate system and generating a preset navigation monitoring route; an unmanned vessel sails along the preset route and collects water depth, heading and attitude data through a single-beam bathymetry system to form a monitoring point cloud data set; mapping the data set to the dock BIM model coordinate system through a seven-parameter coordinate conversion method, and constructing a three-dimensional underwater siltation grid model using a grid interpolation algorithm; based on the model, predicting future siltation trends and obtaining the current warning level using a time series analysis algorithm, and recommending the optimal siltation area to the operation and maintenance end in combination with the actual siltation cost, thereby solving the problems of low efficiency and insufficient precision of traditional methods and significantly improving the safety, economy and data-driven decision-making capabilities of dock monitoring and management.

Description

Technical Field

[0001] The present invention relates to the field of marine engineering and intelligent monitoring technology, and in particular to a method, system and equipment for monitoring underwater siltation of a dock based on a BIM model. Background Art

[0002] Underwater siltation, as a key factor affecting the water depth of the dock and the safety of ship berthing, often requires high-frequency and high-precision monitoring.

[0003] However, traditional dock underwater sedimentation monitoring mainly relies on manually driven small boats equipped with sonar equipment. It is significantly affected by weather conditions such as strong winds and heavy rains. A single operation takes up to several hours and requires manual data analysis, which cannot meet the high-frequency monitoring needs of modern ports. At the same time, manual operation poses safety hazards such as falling into the water and equipment failure. The operation risk is extremely high, especially in harsh environments. The two-dimensional water depth map of the existing technology cannot intuitively reflect the volume and spatial distribution of sedimentation, as well as the spatial relationship with the dock structure, and it is difficult to support the design of a three-dimensional dredging plan. Fixed sonar stations or low-precision sensors have difficulty capturing small-scale sedimentation changes, such as sediments with a particle size of less than 10 cm, and the spacing between survey lines is fixed, which cannot achieve encrypted sampling in areas with high sedimentation risk.

[0004] Furthermore, existing technologies lack a digital infrastructure, making it difficult to integrate multi-source data, such as terminal BIM model data, hydrological data, and equipment operation and maintenance records. This easily creates data silos, making it difficult to assess the impact of siltation on terminal structural safety from a global perspective. Furthermore, manually comparing historical data to determine siltation trends often fails to provide real-time early warnings, which can lead to delayed or excessive desilting operations and increased equipment maintenance costs.

[0005] Therefore, it is necessary to provide a method, system and equipment for monitoring underwater siltation of a dock based on a BIM model to solve the above technical problems. Summary of the Invention

[0006] In order to solve the above technical problems, the present invention provides a method, system and equipment for monitoring underwater siltation at a dock based on a BIM model, which is used to solve the problems of low monitoring efficiency and intelligence level, insufficient accuracy, isolated data and delayed decision-making in traditional methods of monitoring underwater siltation at a dock.

[0007] The present invention provides a method for monitoring underwater siltation at a dock based on a BIM model, the monitoring method comprising:

[0008] Delineate underwater sedimentation monitoring areas and generate preset navigation monitoring routes based on the coordinate system of the dock BIM model;

[0009] The unmanned vessel travels along the preset navigation monitoring route and synchronously collects water depth, heading and attitude data through a single-beam bathymetry system to form a monitoring point cloud data set;

[0010] The monitoring point cloud dataset is mapped to the coordinate system of the dock BIM model using a seven-parameter coordinate transformation method, and a three-dimensional underwater siltation grid model is constructed using a grid interpolation algorithm;

[0011] Based on the three-dimensional underwater siltation grid model, a time series analysis algorithm is used to predict future siltation trends and obtain the current siltation warning level. Combined with the actual siltation cost, the optimal siltation area and project volume are automatically recommended to the operation and maintenance end.

[0012] Preferably, the process of demarcating the underwater sedimentation monitoring area and generating a preset navigation monitoring route based on the coordinate system of the dock BIM model specifically includes:

[0013] Obtaining the terminal BIM model, parsing coordinate data in the terminal BIM model, and establishing a three-dimensional spatial reference system, i.e., a coordinate system of the terminal BIM model;

[0014] The visual interface of the dock BIM model supports manual or automatic selection of the underwater sedimentation monitoring area, and automatically filters out invalid areas in the underwater sedimentation monitoring area in combination with historical water depth data;

[0015] The preset navigation monitoring route is generated based on the underwater sedimentation monitoring area by using the equidistant parallel line method.

[0016] Preferably, the generating of the preset navigation monitoring route based on the underwater sedimentation monitoring area by the equidistant parallel line method specifically includes:

[0017] The underwater sedimentation monitoring area is divided into a regular rectangular monitoring area and a concave-convex shoreline monitoring area. For the regular rectangular monitoring area, the preset navigation monitoring routes are distributed in parallel and back; for the concave-convex shoreline monitoring area, the preset navigation monitoring routes automatically adapt to the boundary contour of the concave-convex shoreline monitoring area;

[0018] The initial survey line spacing of the preset navigation monitoring route is set, and the historical sedimentation database is called to obtain the historical sedimentation rate. For the underwater sedimentation monitoring area where the historical sedimentation rate is greater than the maximum sedimentation rate threshold, the initial survey line spacing is automatically shortened to the minimum survey line spacing threshold; for the area where the historical sedimentation rate is less than the minimum sedimentation rate threshold, the initial survey line spacing is automatically relaxed to the maximum survey line spacing threshold.

[0019] Preferably, the unmanned vessel travels along the preset navigation monitoring route and synchronously collects water depth, heading and attitude data through a single-beam bathymetry system to form a monitoring point cloud data set, specifically including:

[0020] The unmanned vessel travels along the preset navigation monitoring route at a preset navigation speed, automatically identifies surface obstacles and determines obstacle-free areas through an omnidirectional camera device carried by the vessel, triggers an obstacle avoidance algorithm to automatically bypass the obstacle-free areas, and returns to the preset navigation monitoring route to continue traveling after bypassing.

[0021] The unmanned vessel is equipped with the single-beam echo sounding system, and the single-beam echo sounding system includes a transducer, a GNSS receiver and an attitude compass;

[0022] The transducer is used to transmit sonar echoes at a preset transmission frequency and simultaneously collect water depth data; the GNSS receiver is used to output heading data, namely longitude, latitude and altitude data, in real time; the attitude compass is used to monitor attitude data, namely roll, pitch and heading angle data, in real time, and synchronously associate the water depth, heading and attitude data through timestamps to generate an original point cloud dataset carrying spatial attitude information;

[0023] The original point cloud data set is compressed and encrypted by the edge computing node to generate the monitoring point cloud data set.

[0024] Preferably, mapping the monitoring point cloud dataset to the coordinate system of the dock BIM model by a seven-parameter coordinate transformation method, and constructing a three-dimensional underwater siltation grid model by a grid interpolation algorithm, specifically includes:

[0025] The point cloud data corresponding to the blank area of the obstacle is supplemented by using the neighboring point interpolation method, and the noise signal in the sonar echo is eliminated by using the wavelet transform algorithm;

[0026] The WGS-84 coordinate system of the heading data is converted to the coordinate system of the dock BIM model through the seven-parameter coordinate conversion method. The corresponding coordinate system conversion formula is as follows:

[0027] Where (X, Y, Z) represents the coordinates in the coordinate system of the terminal BIM model; is the scale factor; is the rotation parameter; Represents coordinates in the WGS-84 coordinate system; represents the translation parameter;

[0028] Using the grid interpolation algorithm, the three-dimensional underwater siltation grid model is generated based on the monitoring point cloud data set;

[0029] Acquiring the spectral characteristics of the sonar echo and inputting them into a random forest model, automatically distinguishing natural sedimentation elements from artificial waste elements in the three-dimensional underwater sedimentation grid model, and automatically calculating the sedimentation volume, sedimentation thickness and scour boundary position of the natural sedimentation elements;

[0030] A three-color heat map is generated according to the siltation thickness, and is superimposed on the three-dimensional underwater siltation grid model in real time through the API interface of the wharf BIM model to generate a dynamic siltation change twin and send it to the operation and maintenance end. The dynamic siltation change twin supports the operation and maintenance end to view the siltation changes in any area through interactive operations.

[0031] Preferably, the grid interpolation algorithm is used to generate the three-dimensional underwater siltation grid model based on the monitoring point cloud dataset, specifically including:

[0032] Based on the discrete data points i and j in the monitoring point cloud data set and the corresponding water depth data and , i=1,2,…,N, j=1,2,…,N, N is the total number of discrete data points in the monitoring point cloud dataset, and the variance function value is calculated as follows:

[0033] Where, represents the variance function value between discrete data points i and j; d represents the spatial distance between discrete data points i and j; M(d) represents the total number of pairs of discrete data points with a spatial distance of d; Represent the water depth data of discrete data points i and j respectively;

[0034] Set the coordinates of the grid nodes in the three-dimensional underwater sedimentation grid model to , , , where m and n represent grid indexes; Indicates the minimum coordinate value of the three-dimensional underwater sedimentation grid model; represents the grid step size;

[0035] The grid interpolation algorithm is used to calculate the grid nodes The corresponding water depth data as follows:

[0036] Where, Represents the weight coefficient, by solving the linear equations To confirm, Represents the variation function value between discrete data point i and grid node; represents the Lagrange multiplier;

[0037] The grid nodes With the water depth data Correlation to generate three-dimensional underwater sedimentation grid data points , based on all the three-dimensional underwater sedimentation grid data points Construct the three-dimensional underwater sedimentation grid model.

[0038] Preferably, based on the three-dimensional underwater siltation grid model, a time series analysis algorithm is used to predict future siltation trends and obtain the current siltation warning level. Combined with the actual siltation cost, the optimal siltation area and engineering volume are automatically recommended to the operation and maintenance end, specifically including:

[0039] Calculating a current sedimentation rate of each grid unit area in the three-dimensional underwater sedimentation grid model, and identifying an area where the current sedimentation rate is greater than a critical sedimentation rate threshold as a sedimentation warning area;

[0040] For the sedimentation warning area, using a long short-term memory network model to predict the future sedimentation thickness and output the future sedimentation trend;

[0041] A three-level siltation warning mechanism and a siltation thickness warning threshold for each level are set. When the future siltation thickness reaches the siltation thickness warning threshold for the corresponding level, the optimal siltation area and engineering volume are automatically recommended to the operation and maintenance end in combination with the actual siltation cost.

[0042] A dock underwater sedimentation monitoring system based on a BIM model, the monitoring system comprising:

[0043] The route determination module is used to define the underwater sedimentation monitoring area based on the coordinate system of the dock BIM model and generate a preset navigation monitoring route;

[0044] A data acquisition module is used for the unmanned vessel to travel according to the preset navigation monitoring route and synchronously collect water depth, heading and attitude data through a single-beam bathymetry system to form a monitoring point cloud data set;

[0045] A model construction module is used to map the monitoring point cloud dataset to the coordinate system of the dock BIM model through a seven-parameter coordinate transformation method, and to construct a three-dimensional underwater siltation grid model using a grid interpolation algorithm;

[0046] The result recommendation module is used to predict the future sedimentation trend and obtain the current sedimentation warning level based on the three-dimensional underwater sedimentation grid model using a time series analysis algorithm, and automatically recommend the optimal sedimentation area and project volume to the operation and maintenance end in combination with the actual sedimentation cost.

[0047] An electronic device includes a memory and a processor, wherein the memory stores a computer program. When the processor runs the computer program stored in the memory, the processor executes the steps of a method for monitoring underwater siltation of a dock based on a BIM model as described above.

[0048] A readable storage medium having a computer program stored therein, wherein the computer program, when executed by a processor, is used to implement the steps of a method for monitoring underwater siltation of a dock based on a BIM model as described above.

[0049] Compared with related technologies, the method, system and equipment for monitoring underwater siltation at docks based on a BIM model provided by the present invention have the following beneficial effects:

[0050] The present invention demarcates the underwater siltation monitoring area based on the coordinate system of the wharf BIM model and generates a preset navigation monitoring route; the unmanned vessel travels along the preset navigation monitoring route and synchronously collects water depth, heading and attitude data through a single-beam echo sounder system to form a monitoring point cloud data set; the monitoring point cloud data set is mapped to the coordinate system of the wharf BIM model through a seven-parameter coordinate conversion method, and a three-dimensional underwater siltation grid model is constructed using a grid interpolation algorithm; based on the three-dimensional underwater siltation grid model, a time series analysis algorithm is used to predict future siltation trends and obtain the current siltation warning level, and combined with the actual siltation cost, the optimal siltation area and project volume are automatically recommended to the operation and maintenance end, thereby solving the problems of low efficiency and intelligence level, insufficient accuracy, data isolation and decision-making lag in traditional wharf siltation monitoring methods, and significantly improving the safety, economy and data-driven decision-making capabilities of wharf monitoring and management.

[0051] The present invention uses an unmanned vessel equipped with a single-beam bathymetric system to achieve real-time, automated siltation monitoring, with a short single operation time and stable operation in windy and wavey environments, significantly reducing labor costs and operational risks. The method of the present invention can delineate the monitoring area based on the BIM model, dynamically adjust the survey line spacing based on historical siltation data, ensure data coverage integrity and sampling density, and improve the efficiency of siltation monitoring. It also uses a seven-parameter coordinate transformation method to convert the WGS-84 coordinate system into the dock BIM model coordinate system, ensuring spatial alignment of multi-source data. At the same time, it uses a grid interpolation algorithm to generate a three-dimensional grid model, and combines the random forest model to distinguish natural silt from artificial waste, significantly reducing the calculation error of siltation volume and improving the monitoring accuracy of siltation thickness. The present invention predicts future siltation trends through an LSTM time series model, reducing early warning response time, and can identify abnormal siltation risks in advance. In combination with actual siltation costs, such as equipment shift fees, transportation costs, and environmental protection costs, it can automatically recommend the optimal siltation area and project volume, saving equipment costs and avoiding problems such as excessive siltation and delayed maintenance. Furthermore, by constructing a dynamic twin of sedimentation changes, this invention can overlay monitoring data and desilting progress in real time, supporting interactive operations such as 3D sectioning and volume calculation. This enables closed-loop sedimentation management encompassing monitoring, prediction, disposal, and archiving, improving data tracing efficiency. This invention, for the first time, combines BIM models, unmanned vessels, and machine learning algorithms to form a complete technology chain encompassing spatial benchmarking, automated data collection, intelligent analysis, and closed-loop management, driving the digital transformation of sedimentation monitoring in port and navigation projects. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 A flowchart of a method for monitoring underwater siltation at a dock based on a BIM model provided by an embodiment of the present invention;

[0053] Figure 2 A system block diagram of a dock underwater sedimentation monitoring system based on a BIM model provided by an embodiment of the present invention;

[0054] Figure 3 A schematic diagram of the hardware structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0055] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0056] like Figure 1FIG. 1 is a flow chart of a method for monitoring underwater siltation of a dock based on a BIM model according to an embodiment of the present invention. Figure 1 The execution subject of the method shown may be a software and / or hardware device. The execution subject of the present application may include but is not limited to at least one of the following: user equipment, network equipment, etc. Among them, user equipment may include but is not limited to computers, smart phones, personal digital assistants (PDAs) and the electronic devices mentioned above. Network equipment may include but is not limited to a single network server, a server group consisting of multiple network servers, or a cloud based on cloud computing consisting of a large number of computers or network servers, wherein cloud computing is a type of distributed computing, a super virtual computer composed of a group of loosely coupled computers. This embodiment does not limit this. It includes steps S1 to S4, as follows:

[0057] S1, delineate the underwater sedimentation monitoring area based on the coordinate system of the dock BIM model and generate a preset navigation monitoring route;

[0058] The terminal BIM model refers to a three-dimensional digital model of the terminal constructed using parametric modeling technology. This model integrates the terminal's entire lifecycle data, including its geometric dimensions, spatial coordinates, and structural properties, and serves as the core vehicle for delineating monitoring areas and establishing spatial benchmarks. The underwater sedimentation monitoring area refers to the area of the terminal that requires underwater sedimentation monitoring. The pre-set navigation monitoring route refers to the unmanned vessel navigation monitoring route generated based on the underwater sedimentation monitoring area. Its planning process must balance efficiency and accuracy to ensure the comprehensiveness and uniformity of the collected underwater terrain data.

[0059] In the field of wharf engineering, Building Information Modeling (BIM), a digital model that integrates 3D geometric data with engineering information, provides a precise spatial reference for underwater sedimentation monitoring. This approach first establishes a 3D spatial reference system using the BIM model's coordinate system. This coordinate system is typically based on geographic information data from the wharf design phase and includes precise planar coordinates and elevation references.

[0060] In addition, through the visual interactive interface of the BIM model, combined with historical siltation data and waterway safety requirements, the underwater siltation monitoring area can be manually or automatically selected. This area must cover key waters prone to siltation, such as the front of the wharf, the harbor, and the turning section of the waterway, and automatically filter out invalid mudflat areas to ensure the effectiveness of the monitoring range.

[0061] In the route planning phase, for regular rectangular monitoring areas, the equidistant parallel line method is used to generate parallel round-trip preset navigation monitoring routes to ensure uniformity in data collection. For complex terrain areas such as concave and convex coastlines, the route generation algorithm needs to automatically adapt to the coastline contour to avoid blind spots in siltation monitoring. To optimize siltation monitoring efficiency, a historical siltation database can be called to dynamically adjust the spacing between survey lines based on the historical siltation rates of different regions. Specifically, for high-risk areas where the historical siltation rate is higher than the threshold, the spacing between survey lines can be increased to obtain a higher density of monitoring data; for low-risk areas with stable siltation, the spacing between survey lines can be relaxed to improve monitoring efficiency while ensuring monitoring accuracy.

[0062] S2, the unmanned vessel travels along the preset navigation monitoring route and synchronously collects water depth, heading and attitude data through a single-beam bathymetry system to form a monitoring point cloud dataset;

[0063] It's understandable that an unmanned vessel is an intelligent water-based device equipped with sensors and communication modules. It features autonomous navigation, data collection, and remote control, and is used to replace manned boats in underwater monitoring missions. A single-beam echo sounder system is a device combination that uses sonar technology to measure water depth. It includes a transducer, a GNSS receiver, and an attitude compass. The transducer transmits sonar pulses and receives echoes, calculating water depth using time difference. The GNSS receiver provides the unmanned vessel's heading data, and the attitude compass monitors the vessel's attitude data. The monitoring point cloud dataset is a collection of three-dimensional discrete data collected synchronously by the unmanned vessel. It contains water depth, position, and attitude information and serves as the raw data for underwater terrain modeling.

[0064] It should be noted that the unmanned vessel, serving as the core vehicle for underwater data collection, utilizes a carbon fiber trimaran design, capable of withstanding level 3 wind and wave conditions and stable operation in complex hydrological environments. The unmanned vessel autonomously navigates along a pre-set route at a pre-set speed. Using its onboard omnidirectional cameras, the vessel identifies surface obstacles, such as floating objects and operating vessels, in real time. This triggers an obstacle avoidance algorithm to automatically circumvent areas where the unmanned vessel has actively circumvented data collection. Upon completion, the vessel automatically returns to its original route, ensuring the continuity and integrity of the data collection path.

[0065] The single-beam bathymetry system onboard the unmanned vessel utilizes multiple sensors working in concert. Specifically, a high-precision transducer emits sonar pulses at a preset frequency, calculating water depth by measuring the round-trip time of the sound waves. A GNSS receiver provides real-time data on the vessel's longitude, latitude, and altitude. An attitude compass continuously monitors the vessel's roll, pitch, and heading. Data from these three sensors is synchronized via hardware timestamping, forming a monitoring point cloud dataset containing spatial position, water depth, and attitude information.

[0066] S3, mapping the monitoring point cloud dataset to the coordinate system of the wharf BIM model through a seven-parameter coordinate transformation method, and constructing a three-dimensional underwater siltation grid model using a grid interpolation algorithm;

[0067] Among them, the seven-parameter coordinate transformation method is a mathematical method for coordinate transformation between different coordinate systems, which can transform the WGS-84 coordinate system into the BIM model coordinate system by translation parameters. , rotation parameters and scale factor A total of seven parameters can achieve coordinate system unity and control errors to the centimeter level. The grid interpolation algorithm converts discrete point cloud data into regular grid data. It evaluates spatial correlation by calculating the variance function value, and then generates a high-precision three-dimensional grid model to intuitively display the undulations of the underwater terrain. The three-dimensional underwater siltation grid model refers to a high-precision three-dimensional terrain model constructed based on monitoring point cloud data. It uses a grid interpolation algorithm to convert discrete points into a regular grid to intuitively display the undulations and siltation distribution of the underwater terrain.

[0068] In practical applications, the WGS-84 coordinate system data collected by the GNSS receiver can be converted into the coordinate system of the terminal BIM model through the seven-parameter coordinate conversion method. , rotation parameters and scale factor A total of seven parameters can be used to eliminate systematic deviations between different coordinate systems and achieve spatial datum unification between monitoring data and BIM models. To address the problem of missing data in blank areas caused by obstacle avoidance by unmanned boats, the neighboring point interpolation method can be used to supplement the discrete point cloud data, and the wavelet transform algorithm can be used to remove noise signals in sonar echoes to improve data quality.

[0069] During the 3D modeling phase, a gridded interpolation algorithm is used to convert discrete point cloud data into a regular grid model. This algorithm evaluates the correlation between data points by calculating the spatial variation function value, assigns a weight coefficient to each grid node based on the distance attenuation law, and generates a 3D underwater sedimentation grid model with a preset resolution. Once the model is constructed, the spectral characteristics of the sonar echo can be automatically extracted and input into a pre-trained random forest model to automatically classify natural sediments and artificial waste, thereby improving classification accuracy. At the same time, the sedimentation volume, thickness distribution, and scour boundary location of each area can be calculated based on the grid model, generating a red, yellow, and green heat map. This map can be superimposed on the 3D scene in real time through the BIM model's application program interface, forming a dynamic sedimentation change twin. This also allows operations and maintenance personnel to intuitively view sedimentation details in any area through interactive operations such as 3D sectioning and data query.

[0070] S4, based on the three-dimensional underwater siltation grid model, uses a time series analysis algorithm to predict future siltation trends and obtain the current siltation warning level. Combined with the actual siltation cost, it automatically recommends the optimal siltation area and project volume to the operation and maintenance end.

[0071] In practical applications, the time series analysis algorithm is an algorithm that predicts future siltation trends based on time series data. It uses a long short-term memory (LSTM) model to train historical siltation data to predict future changes in siltation thickness and output the direction and extent of future siltation. The current siltation warning level is determined based on real-time monitoring of siltation thickness and predicted trends, and is used to trigger appropriate operational and maintenance responses. The actual desilting cost refers to the cost required to complete desilting operations in a specific area and is used to optimize the economic efficiency of desilting solutions. The optimal desilting area is the dock desilting area with the highest desilting efficiency and lowest cost.

[0072] Based on the constructed three-dimensional underwater siltation grid model, historical monitoring data can be trained through a time series analysis algorithm to predict future siltation trends. The algorithm learns the long-term dependencies in the time series, identifies the changing patterns of siltation rates, and outputs a forecast report on future siltation trends, clarifying the spatial distribution and growth rate of future siltation thickness. At the same time, a three-level siltation warning mechanism can be set up, with different siltation warning levels corresponding to different siltation response priorities. When the predicted siltation thickness reaches 0.5 meters, a yellow warning is triggered, prompting increased monitoring frequency; when the predicted siltation thickness reaches 0.8 meters, an orange warning is triggered, suggesting the start of siltation equipment scheduling; when the predicted siltation thickness reaches 1.0 meters, a red warning is triggered, forcing the emergency siltation process to be initiated.

[0073] When optimizing dredging plans, the system automatically recommends optimal dredging areas and project quantities, taking into account actual dredging costs, equipment shift fees, transportation costs, environmental protection treatment fees, and other cost factors. 3D model integration improves the accuracy of dredging project quantities and supports real-time tracking of dredging progress. Upon completion, the BIM model is automatically updated, forming a closed-loop management system for monitoring, analysis, decision-making, and feedback throughout the entire lifecycle.

[0074] In the specific implementation process, the underwater sedimentation monitoring area is delineated based on the coordinate system of the dock BIM model and a preset navigation monitoring route is generated, specifically including:

[0075] Obtaining the terminal BIM model, parsing coordinate data in the terminal BIM model, and establishing a three-dimensional spatial reference system, i.e., a coordinate system of the terminal BIM model;

[0076] The visual interface of the dock BIM model supports manual or automatic selection of the underwater sedimentation monitoring area, and automatically filters out invalid areas in the underwater sedimentation monitoring area in combination with historical water depth data;

[0077] The preset navigation monitoring route is generated based on the underwater sedimentation monitoring area by using the equidistant parallel line method.

[0078] The method of generating the preset navigation monitoring route based on the underwater sedimentation monitoring area by using the equidistant parallel line method specifically includes:

[0079] The underwater sedimentation monitoring area is divided into a regular rectangular monitoring area and a concave-convex shoreline monitoring area. For the regular rectangular monitoring area, the preset navigation monitoring routes are distributed in parallel and back; for the concave-convex shoreline monitoring area, the preset navigation monitoring routes automatically adapt to the boundary contour of the concave-convex shoreline monitoring area;

[0080] The initial survey line spacing of the preset navigation monitoring route is set, and the historical sedimentation database is called to obtain the historical sedimentation rate. For the underwater sedimentation monitoring area where the historical sedimentation rate is greater than the maximum sedimentation rate threshold, the initial survey line spacing is automatically shortened to the minimum survey line spacing threshold; for the area where the historical sedimentation rate is less than the minimum sedimentation rate threshold, the initial survey line spacing is automatically relaxed to the maximum survey line spacing threshold.

[0081] In practical applications, the first step is to obtain the terminal building information model, which integrates the 3D geometric data from the terminal design phase, such as the terminal platform, pile foundations, and shoreline contours, as well as geographic information data such as the coordinate system and elevation datum. By parsing the coordinate data in the model, a 3D spatial reference system with the terminal's local coordinate system as its core is established, which is the coordinate system of the terminal BIM model.

[0082] Secondly, in the visual operation interface of the terminal BIM model, underwater siltation monitoring areas can be selected through manual interaction or automated algorithms based on monitoring needs. In manual selection mode, operations and maintenance personnel use their experience to outline the boundaries of siltation-prone areas such as the terminal front, harbor, and channel bends in the three-dimensional scene; the automatic selection mode uses an algorithm to identify areas in the model with water depths less than a preset depth value, or combines historical siltation distribution data to automatically generate a monitoring range. During the selection process, the historical water depth database can be simultaneously called to automatically filter out mudflats or other non-navigable waters with a water depth of less than 0.15 meters, eliminating invalid monitoring areas and focusing on waters that have a real impact on the safety of terminal operations. For example, a container terminal used the automatic selection function to limit the monitoring range to waters within 200 meters of the terminal front and with a water depth greater than 5 meters, while excluding the shallow areas on the east side, thereby improving monitoring efficiency.

[0083] Finally, based on the designated underwater sedimentation monitoring area, the equidistant parallel line method can be used to generate preset navigation monitoring routes. This method divides the monitoring area into regular grid cells and covers the entire area with parallel routes. For rectangular monitoring areas with regular shapes, the routes are distributed in parallel and the default line spacing is 2 meters to ensure uniform data collection density; for complex terrain areas such as concave and convex coastlines, irregular islands and reefs, the route generation algorithm automatically adapts to the regional boundary contours and avoids monitoring blind spots while ensuring route continuity through segmented straight line fitting. For example, in the concave and convex coastline area of a certain estuary wharf, an "S"-shaped detour route can be automatically generated to ensure that all data within 5 meters of the coastline are covered.

[0084] In addition, historical sedimentation rate data can be combined during the route generation process. For high-risk areas, such as areas with a historical sedimentation rate greater than 0.2 meters / month, the survey line spacing can be automatically increased to 0.5 meters. For low-risk areas, such as areas with a historical sedimentation rate less than 0.05 meters / month, the survey line spacing can be automatically relaxed to 5 meters. In this way, differentiated sedimentation monitoring strategies can be formulated to optimize the operation path of the unmanned ship and shorten the operation time while ensuring monitoring accuracy.

[0085] This process leverages the BIM model's spatial benchmarking, the visual interface's human-machine collaboration capabilities, and algorithm-driven route optimization strategies to achieve scientific monitoring area delineation and efficient route planning. This reduces the subjectivity and errors of manual planning, laying the foundation for subsequent unmanned vessel data collection, 3D modeling, and intelligent analysis. Furthermore, the deep integration of historical data and spatial models enhances the monitoring solution's relevance and engineering adaptability, making it particularly suitable for port waters with complex terrain and variable siltation patterns.

[0086] The unmanned vessel travels along the preset navigation monitoring route and synchronously collects water depth, heading and attitude data through a single-beam bathymetry system to form a monitoring point cloud data set, specifically including:

[0087] The unmanned vessel travels along the preset navigation monitoring route at a preset navigation speed, automatically identifies surface obstacles and determines obstacle-free areas through an omnidirectional camera device carried by the vessel, triggers an obstacle avoidance algorithm to automatically bypass the obstacle-free areas, and returns to the preset navigation monitoring route to continue traveling after bypassing.

[0088] The unmanned vessel is equipped with the single-beam echo sounding system, and the single-beam echo sounding system includes a transducer, a GNSS receiver and an attitude compass;

[0089] The transducer is used to transmit sonar echoes at a preset transmission frequency and simultaneously collect water depth data; the GNSS receiver is used to output heading data, namely longitude, latitude and altitude data, in real time; the attitude compass is used to monitor attitude data, namely roll, pitch and heading angle data, in real time, and synchronously associate the water depth, heading and attitude data through timestamps to generate an original point cloud dataset carrying spatial attitude information;

[0090] The original point cloud data set is compressed and encrypted by the edge computing node to generate the monitoring point cloud data set.

[0091] The unmanned vessel travels steadily along a pre-set navigation monitoring route at a speed of 2 meters per second. During navigation, an onboard omnidirectional camera, such as a high-definition camera or lidar, continuously scans the water surface. Using image recognition algorithms, it detects surface obstacles such as floating objects, operating vessels, and water-based facilities in real time. When an obstacle is detected, it automatically demarcates the clear area around the obstacle and triggers obstacle avoidance algorithms, such as the A* path planning algorithm, to automatically generate a detour route, allowing the unmanned vessel to avoid the obstacle at a safe distance, typically 5-10 meters. After the detour is completed, the unmanned vessel returns to the pre-set route using GNSS differential positioning technology and resumes data collection. This process ensures safe operation and route continuity for the unmanned vessel in complex water environments. For example, during monitoring of a container terminal, the unmanned vessel successfully avoided three floating obstacles with a route regression error of less than 0.5 meters.

[0092] The single-beam depth sounding system carried by the unmanned vessel consists of three core sensors: a transducer, a GNSS receiver, and an attitude compass. Specifically, the transducer emits sonar pulses underwater at a preset transmission frequency, such as 200 kHz. By receiving the echo signal reflected from the bottom of the water, it calculates the round-trip time of the sound waves to obtain water depth data. Its depth measurement accuracy can reach the centimeter level and is applicable to water depths ranging from 0.15 to 300 meters. The GNSS receiver outputs the unmanned vessel's heading data, namely longitude, latitude, and altitude data, in real time, with positioning accuracy reaching the millimeter level, providing a precise plane and elevation reference for water depth data. The attitude compass continuously monitors the unmanned vessel's attitude data, namely roll angle, pitch angle, and heading angle, with an accuracy of ±0.1 degrees, and is used to correct the impact of changes in the hull's attitude on water depth measurements. The three types of sensors achieve data synchronization through hardware timestamps, associating water depth values with corresponding spatial coordinates and attitude parameters to form an original point cloud dataset with complete spatiotemporal attitude information. For example, each set of data contains the water depth value, longitude, latitude, altitude and three attitude angle parameters at a certain moment to ensure the spatial consistency of the data.

[0093] After the raw point cloud dataset is generated, it can be compressed and encrypted by edge computing nodes to generate a monitoring point cloud dataset. First, efficient compression algorithms, such as wavelet compression, can be used to reduce the data volume of the raw point cloud dataset, with a compression ratio of up to 10:1 to adapt to network transmission bandwidth. At the same time, encryption algorithms can be used to encrypt the data to ensure data security during transmission. The monitoring point cloud dataset can then be transmitted to a shore-based server in real time, with transmission rates reaching hundreds of megabits and latency less than 100 milliseconds to meet real-time monitoring needs. For example, during monitoring at a dock at the Yangtze River Estuary, an unmanned vessel generates approximately 10GB of raw data per hour, which is compressed and encrypted and transmitted to the server in real time, providing efficient and reliable data support for subsequent 3D modeling.

[0094] Through the above method, the automated and high-precision collection of underwater terrain data can be achieved, effectively solving the efficiency bottleneck and data synchronization problems of traditional manual monitoring, and laying a solid foundation for real-time monitoring and dynamic analysis of underwater siltation at the dock.

[0095] The monitoring point cloud dataset is mapped to the coordinate system of the dock BIM model by a seven-parameter coordinate transformation method, and a three-dimensional underwater siltation grid model is constructed using a grid interpolation algorithm, specifically including:

[0096] The point cloud data corresponding to the blank area of the obstacle is supplemented by using the neighboring point interpolation method, and the noise signal in the sonar echo is eliminated by using the wavelet transform algorithm;

[0097] The WGS-84 coordinate system of the heading data is converted to the coordinate system of the dock BIM model through the seven-parameter coordinate conversion method. The corresponding coordinate system conversion formula is as follows:

[0098] Where (X, Y, Z) represents the coordinates in the coordinate system of the terminal BIM model; is the scale factor; is the rotation parameter; Represents coordinates in the WGS-84 coordinate system; represents the translation parameter;

[0099] Using the grid interpolation algorithm, the three-dimensional underwater siltation grid model is generated based on the monitoring point cloud data set;

[0100] Acquiring the spectral characteristics of the sonar echo and inputting them into a random forest model, automatically distinguishing natural sedimentation elements from artificial waste elements in the three-dimensional underwater sedimentation grid model, and automatically calculating the sedimentation volume, sedimentation thickness and scour boundary position of the natural sedimentation elements;

[0101] A three-color heat map is generated according to the siltation thickness, and is superimposed on the three-dimensional underwater siltation grid model in real time through the API interface of the wharf BIM model to generate a dynamic siltation change twin and send it to the operation and maintenance end. The dynamic siltation change twin supports the operation and maintenance end to view the siltation changes in any area through interactive operations.

[0102] In practical applications, the neighboring point interpolation method can be used to complete the data for the blank areas caused by obstacle avoidance caused by unmanned boats. This method is based on the principle of correlation of spatial data, and uses the water depth values of valid data points around the blank area to estimate the missing point data through a weighted average algorithm. For example, for the blank area corresponding to a certain obstacle, 10 valid data points within a range of 5 meters around it are selected, and weights are assigned according to the distance to calculate the water depth estimate of the missing point to ensure the integrity of the data. At the same time, the wavelet transform algorithm can be used to reduce the noise of the sonar echo data, and high-frequency noise signals such as hull vibration interference and water turbulence noise can be eliminated through multi-scale analysis to improve the signal-to-noise ratio of the water depth data and provide a high-quality data source for subsequent modeling.

[0103] Secondly, the spatial reference of the monitoring data can be unified through the seven-parameter coordinate transformation method. This method transforms the WGS-84 global coordinate system of the heading data into the coordinate system of the terminal BIM model through seven parameters, namely translation, rotation and scale transformation. Specifically, the translation parameter Used to adjust the coordinate origin deviation and rotation parameters Used to correct the tilt of the coordinate axis, scale factor Used to compensate for differences in coordinate system scale. For example, a dock BIM model uses a local engineering coordinate system. After seven-parameter conversion, the planar error of GNSS data is reduced from meters to within 5 centimeters, meeting the accuracy requirements of 3D modeling.

[0104] During the 3D modeling phase, gridded interpolation algorithms, such as kriging, are used to convert discrete point cloud data into a regular grid model. This algorithm first calculates the variogram between data points to assess spatial autocorrelation. Then, based on the distance decay law, it assigns weights to each grid node, generating a 3D underwater sedimentation grid model with a resolution of 0.1 m x 0.1 m. For example, in the waters off the coast of a particular dock, an interpolation algorithm was used to convert 20,000 discrete points into 2 million grid nodes, accurately capturing the subtle undulations of the underwater terrain.

[0105] Once the model is built, the spectral characteristics of the sonar echoes, such as the dominant frequency, bandwidth, and attenuation rate, can be extracted and fed into a pre-trained random forest model to automatically classify natural sedimentation elements from artificial waste elements. Once classified, the volume, thickness distribution, and scour boundary location of natural sedimentation elements can be automatically calculated based on the grid model. For example, the sedimentation volume of a specific area can be calculated as 2,300 cubic meters using the integration method.

[0106] Finally, a red, yellow, and green heat map can be generated based on the sediment thickness threshold. This can be overlaid on the 3D mesh model in real time through the BIM model's application programming interface (API), forming a dynamic twin of sediment changes. Operations and maintenance personnel can then view sedimentation details in any area through interactive operations such as 3D sectioning and data querying. For example, by dragging the mouse, a cross-section perpendicular to the shoreline can be generated, visually displaying changes in sediment thickness at different depths. This dynamic model also supports real-time data updates, automatically synchronizing the latest monitoring data after dredging is completed, forming a full-cycle, traceable digital management scenario.

[0107] Through the organic combination of data completion, coordinate transformation, intelligent classification and visual presentation, efficient conversion from original monitoring data to three-dimensional decision-making models is achieved, providing strong technical support for accurate assessment and scientific decision-making of wharf siltation.

[0108] The method of using the grid interpolation algorithm to generate the three-dimensional underwater siltation grid model based on the monitoring point cloud dataset specifically includes:

[0109] Based on the discrete data points i and j in the monitoring point cloud data set and the corresponding water depth data and , i=1,2,…,N, j=1,2,…,N, N is the total number of discrete data points in the monitoring point cloud dataset, and the variance function value is calculated as follows:

[0110] Where, represents the variance function value between discrete data points i and j; d represents the spatial distance between discrete data points i and j; M(d) represents the total number of pairs of discrete data points with a spatial distance of d; Represent the water depth data of discrete data points i and j respectively;

[0111] Set the coordinates of the grid nodes in the three-dimensional underwater sedimentation grid model to , , , where m and n represent grid indexes; Indicates the minimum coordinate value of the three-dimensional underwater sedimentation grid model; represents the grid step size;

[0112] The grid interpolation algorithm is used to calculate the grid nodes The corresponding water depth data as follows:

[0113] Where, Represents the weight coefficient, by solving the linear equations To confirm, Represents the variation function value between discrete data point i and grid node; represents the Lagrange multiplier;

[0114] The grid nodes With the water depth data Correlation to generate three-dimensional underwater sedimentation grid data points , based on all the three-dimensional underwater sedimentation grid data points Construct the three-dimensional underwater siltation grid model.

[0115] When constructing a three-dimensional underwater sedimentation grid model, the grid interpolation algorithm can achieve the conversion from discrete monitoring data to a continuous grid model by quantifying the spatial correlation of data points.

[0116] First, based on the discrete data points in the monitoring point cloud dataset and their corresponding water depth values, the variogram value can be calculated to evaluate spatial autocorrelation. Specifically, the variogram reveals the spatial distribution characteristics of the data by measuring the change pattern of the water depth difference between any two points as the spatial distance changes. For each pair of discrete data points, their spatial distance can be calculated, and all point pairs at the same distance can be counted, and the variogram value can be calculated using the corresponding formula. This value reflects the degree of water depth difference between the two points. If the water depth difference is small, it indicates that the spatial correlation between the point pairs is strong, and as the distance between the point pairs increases, the variogram value tends to stabilize and the spatial correlation between the point pairs weakens. For example, in the monitoring data of a certain dock, the variogram value of point pairs with a distance of less than 5 meters rises rapidly with increasing distance, while the variogram value of point pairs with a distance greater than 20 meters tends to be stable, indicating that the effective spatial range of the area is about 20 meters.

[0117] Next, you can set the grid node coordinates for the 3D underwater sedimentation grid model. Starting from the minimum coordinate value of the monitoring area, a regular grid is generated according to a preset grid resolution, such as 0.1 meters. The plane coordinates of each grid node are obtained by multiplying the grid index by the resolution, forming a regular grid array covering the entire monitoring area. For example, if the minimum coordinates of the monitoring area are (1000 meters, 2000 meters) and the resolution is 0.1 meters, the coordinates of the 10th grid node are (1001 meters, 2001 meters).

[0118] Then, a gridded interpolation algorithm can be used to calculate the water depth value for each grid node. Determining the weight coefficients is the core of the gridded interpolation algorithm and is achieved by solving a system of linear equations. This system of equations uses the variogram value as input to ensure that the interpolation result is unbiased and optimal: on the one hand, the weight of each discrete data point is inversely proportional to the variogram value of its connection to the grid node; on the other hand, the sum of the weights must be 1 to ensure the unbiased interpolation. By solving this system of equations, the water depth value of each grid node is obtained by the weighted sum of the water depth values of all discrete data points. For example, if the weights of the five discrete points surrounding a grid node are 0.2, 0.3, 0.25, 0.15, and 0.1, respectively, the water depth value of the grid node is the cumulative value of the water depth values of each discrete point multiplied by the corresponding weights.

[0119] Finally, the coordinates of all grid nodes can be associated with the calculated water depth values to generate three-dimensional underwater siltation grid data points containing spatial location and attribute information. These data points are arranged in an orderly manner according to the grid index, forming a complete three-dimensional underwater siltation grid model. This model intuitively presents the undulations of the underwater terrain in the form of a regular grid, supporting subsequent siltation volume calculation, thickness distribution analysis, and visualization. For example, this model can quickly locate areas with siltation thickness exceeding 1 meter, and then calculate the siltation volume and generate a silt removal bill of quantities, providing accurate data support for terminal operations and maintenance.

[0120] This process converts sparse discrete monitoring data into a high-density continuous grid model through the organic combination of variation function analysis, grid division, weight solution and data association, effectively improving the accuracy and visualization level of underwater sedimentation monitoring. It is a key technical link in realizing intelligent dredging decision-making.

[0121] Based on the three-dimensional underwater siltation grid model, the time series analysis algorithm is used to predict future siltation trends and obtain the current siltation warning level. Combined with the actual siltation cost, the optimal siltation area and engineering volume are automatically recommended to the operation and maintenance end. Specifically, the following steps are performed:

[0122] Calculating a current sedimentation rate of each grid unit area in the three-dimensional underwater sedimentation grid model, and identifying an area where the current sedimentation rate is greater than a critical sedimentation rate threshold as a sedimentation warning area;

[0123] For the sedimentation warning area, using a long short-term memory network model to predict the future sedimentation thickness and output the future sedimentation trend;

[0124] A three-level siltation warning mechanism and a siltation thickness warning threshold for each level are set. When the future siltation thickness reaches the siltation thickness warning threshold for the corresponding level, the optimal siltation area and engineering volume are automatically recommended to the operation and maintenance end in combination with the actual siltation cost.

[0125] In practical applications, for each grid unit area in a three-dimensional underwater sedimentation grid model, by comparing the monitoring data of two adjacent grid models, such as the monitoring data of a grid model with a seven-day interval, the ratio of the change in sedimentation thickness of each grid unit to the time interval can be calculated to obtain the current sedimentation rate. For example, if the sedimentation thickness of a grid unit area increases from 0.3 meters to 0.5 meters within two weeks, the current sedimentation rate is 0.1 meters per week. In addition, a critical sedimentation rate threshold, such as 0.05 meters per week, can be preset in advance, and areas above this threshold will be identified as sedimentation warning areas. These areas require special attention and should be included in the prediction model due to their rapid sedimentation rate.

[0126] For identified sedimentation warning areas, a long short-term memory (LSTM) model can be used to predict future sedimentation thickness. This model, a type of recurrent neural network, excels at processing long-term dependencies in time series data. By inputting a historical sedimentation rate sequence, such as data from the past 12 weeks, the model automatically learns the temporal patterns of sedimentation changes and outputs a 30-day sedimentation thickness forecast and trend curve. For example, if the sedimentation rates in a certain sedimentation warning area over the past three months were 0.1 meters / week, 0.12 meters / week, and 0.11 meters / week, respectively, the LSTM model predicts that the sedimentation rate will stabilize at around 0.11 meters / week over the next four weeks, and the sedimentation thickness will reach 0.8 meters in 30 days.

[0127] To achieve risk tiered management, a three-level sedimentation warning mechanism can be set up. Different sedimentation warning levels correspond to different desilting response priorities. Specifically, when the predicted sedimentation thickness reaches 0.5 meters, a yellow warning is triggered, prompting increased monitoring frequency; when the predicted sedimentation thickness reaches 0.8 meters, an orange warning is triggered, suggesting the start of desilting equipment scheduling; when the predicted sedimentation thickness reaches 1.0 meters, a red warning is triggered, forcing the start of the emergency desilting process. For example, if the sedimentation thickness in a certain area is predicted to be 0.9 meters in 30 days, reaching the orange warning threshold, it is recommended to start desilting equipment scheduling, and a desilting task work order is generated and pushed to the operation and maintenance personnel.

[0128] When recommending dredging plans, a multi-objective optimization approach can be performed based on actual dredging costs. Actual dredging costs include factors such as equipment shift fees, transportation costs, and environmental treatment fees. The optimal dredging equipment and processes are automatically matched to the thickness, distribution, and type of silt in different warning levels. For example, for yellow warning areas with large shallow areas, efficient and low-cost cutter suction vessels are recommended; for red warning areas containing metal waste, grab vessels are prioritized to avoid equipment damage. Furthermore, the dredging workload can be calculated based on the 3D model, generating a detailed list that includes the area location, silt thickness, volume to be cleared, and a cost estimate.

[0129] Through the above methods, traditional dredging decisions that rely on manual experience can be upgraded to a data-driven scientific dredging process, thereby shortening the decision-making cycle. Moreover, through accurate cost accounting and equipment matching, dredging costs can be greatly reduced, significantly improving the intelligence level and economic benefits of terminal operations and maintenance.

[0130] like Figure 2 FIG. 1 is a system block diagram of a BIM-based dock underwater sedimentation monitoring system according to an embodiment of the present invention. The monitoring system includes:

[0131] The route determination module is used to define the underwater sedimentation monitoring area based on the coordinate system of the dock BIM model and generate a preset navigation monitoring route;

[0132] A data acquisition module is used for the unmanned vessel to travel according to the preset navigation monitoring route and synchronously collect water depth, heading and attitude data through a single-beam bathymetry system to form a monitoring point cloud data set;

[0133] A model construction module is used to map the monitoring point cloud dataset to the coordinate system of the dock BIM model through a seven-parameter coordinate transformation method, and to construct a three-dimensional underwater siltation grid model using a grid interpolation algorithm;

[0134] The result recommendation module is used to predict the future sedimentation trend and obtain the current sedimentation warning level based on the three-dimensional underwater sedimentation grid model using a time series analysis algorithm, and automatically recommend the optimal sedimentation area and project volume to the operation and maintenance end in combination with the actual sedimentation cost.

[0135] Figure 2 The apparatus of the embodiment shown can be used to perform Figure 1 The implementation principles and technical effects of the steps in the method embodiment shown are similar and will not be repeated here.

[0136] An electronic device includes a memory and a processor, wherein the memory stores a computer program. When the processor runs the computer program stored in the memory, the processor executes the steps of a method for monitoring underwater siltation of a dock based on a BIM model as described above.

[0137] like Figure 3 FIG. 1 is a schematic diagram of the hardware structure of an electronic device provided by an embodiment of the present invention. The electronic device 30 includes: a processor 31, a memory 32 and a computer program;

[0138] The memory 32 is used to store the computer program, which may also be a flash memory. The computer program is, for example, an application program or a functional module for implementing the above method.

[0139] The processor 31 is configured to execute the computer program stored in the memory to implement the various steps performed by the device in the above method. For details, please refer to the relevant description in the above method embodiment.

[0140] Optionally, the memory 32 may be independent or integrated with the processor 31 .

[0141] When the memory 32 is a device independent of the processor 31, the device may further include:

[0142] The bus 33 is used to connect the memory 32 and the processor 31 .

[0143] A readable storage medium having a computer program stored therein, wherein the computer program, when executed by a processor, is used to implement the steps of a method for monitoring underwater siltation of a dock based on a BIM model as described above.

[0144] The readable storage medium may be a computer storage medium or a communication medium. Communication media include any medium that facilitates the transfer of computer programs from one location to another. Computer storage media may be any available medium that can be accessed by a general-purpose or special-purpose computer. For example, a readable storage medium is coupled to a processor, enabling the processor to read information from and write information to the readable storage medium. Of course, the readable storage medium may also be an integral part of the processor. The processor and the readable storage medium may be located in an application-specific integrated circuit (ASIC). In addition, the ASIC may be located in a user device. Of course, the processor and the readable storage medium may also exist as discrete components in a communication device. The readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, and the like.

[0145] The present invention also provides a program product, which includes execution instructions stored in a readable storage medium. At least one processor of a device can read the execution instructions from the readable storage medium, and at least one processor executes the execution instructions so that the device implements the methods provided in the various embodiments described above.

[0146] In the embodiments of the above-mentioned devices, it should be understood that the processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASICs), etc. The general-purpose processor may be a microprocessor or any conventional processor. The steps of the method disclosed in the present invention may be directly executed by a hardware processor or by a combination of hardware and software modules within the processor.

[0147] Through the introduction of the above embodiments, the present invention uses a method, system and equipment for monitoring underwater siltation of a dock based on a BIM model. The underwater siltation monitoring area is delineated and a preset navigation monitoring route is generated based on the coordinate system of the dock BIM model. The unmanned vessel travels along the preset navigation monitoring route and synchronously collects water depth, heading and attitude data through a single-beam echo sounder system to form a monitoring point cloud data set. The monitoring point cloud data set is mapped to the coordinate system of the dock BIM model through a seven-parameter coordinate transformation method, and a three-dimensional underwater siltation grid model is constructed using a grid interpolation algorithm. Based on the three-dimensional underwater siltation grid model, a time series analysis algorithm is used to predict future siltation trends and obtain the current siltation warning level. Combined with the actual siltation cost, the optimal siltation area and engineering volume are automatically recommended to the operation and maintenance end, thereby solving the problems of low efficiency and intelligence level, insufficient accuracy, data isolation and decision lag in traditional dock siltation monitoring methods, and significantly improving the safety, economy and data-driven decision-making capabilities of dock monitoring and management.

[0148] The present invention uses an unmanned vessel equipped with a single-beam bathymetric system to achieve real-time, automated siltation monitoring, with a short single operation time. It can operate stably in windy and wave environments, significantly reducing labor costs and operational risks. The method of the present invention can delineate the monitoring area based on the BIM model, dynamically adjust the survey line spacing based on historical siltation data, ensure data coverage integrity and sampling density, and improve siltation monitoring efficiency. It also uses a seven-parameter coordinate conversion method to convert the WGS-84 coordinate system collected by the GNSS receiver into the dock BIM model coordinate system, ensuring spatial alignment of multi-source data. At the same time, it uses a grid interpolation algorithm to generate a three-dimensional grid model, combined with a random forest model to distinguish between natural silt and artificial waste, significantly reducing the calculation error of siltation volume and improving the monitoring accuracy of siltation thickness. The present invention sets a three-level sedimentation early warning mechanism and predicts future sedimentation trends through the LSTM time series model, thereby reducing the early warning response time, identifying abnormal sedimentation risks in advance, and automatically recommending the optimal siltation area and engineering volume based on actual siltation costs, such as equipment shift fees, transportation fees, environmental protection costs, etc., saving equipment costs and avoiding excessive siltation or maintenance delays. In addition, the present invention constructs a dynamic siltation change twin, which can superimpose monitoring data and siltation progress in real time, supports interactive operations such as three-dimensional sectioning and volume calculation, and realizes closed-loop siltation management of monitoring, prediction, disposal and archiving, thereby improving data tracing efficiency. The present invention combines BIM models, unmanned ships and machine learning algorithms for the first time, forming a complete technical chain of spatial benchmarking, automatic collection, intelligent analysis and closed-loop management, and promoting the digital transformation of siltation monitoring in port and shipping projects.

[0149] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the above embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for monitoring underwater siltation at a dock based on a BIM model, characterized in that: The monitoring method comprises: Delineate underwater sedimentation monitoring areas and generate preset navigation monitoring routes based on the coordinate system of the dock BIM model; The unmanned vessel travels along the preset navigation monitoring route and synchronously collects water depth, heading and attitude data through a single-beam bathymetry system to form a monitoring point cloud data set; The monitoring point cloud dataset is mapped to the coordinate system of the dock BIM model using a seven-parameter coordinate transformation method, and a three-dimensional underwater siltation grid model is constructed using a grid interpolation algorithm; Based on the three-dimensional underwater siltation grid model, a time series analysis algorithm is used to predict future siltation trends and obtain the current siltation warning level. Combined with the actual siltation cost, the optimal siltation area and project volume are automatically recommended to the operation and maintenance end; The unmanned vessel travels along the preset navigation monitoring route and synchronously collects water depth, heading and attitude data through a single-beam bathymetry system to form a monitoring point cloud data set, specifically including: The unmanned vessel travels along the preset navigation monitoring route at a preset navigation speed, automatically identifies surface obstacles and determines obstacle-free areas through an omnidirectional camera device carried by the vessel, triggers an obstacle avoidance algorithm to automatically bypass the obstacle-free areas, and returns to the preset navigation monitoring route to continue traveling after bypassing. The unmanned vessel is equipped with the single-beam echo sounding system, and the single-beam echo sounding system includes a transducer, a GNSS receiver and an attitude compass; The transducer is used to transmit sonar echoes at a preset transmission frequency and simultaneously collect water depth data; the GNSS receiver is used to output heading data, namely longitude, latitude and altitude data, in real time; the attitude compass is used to monitor attitude data, namely roll, pitch and heading angle data, in real time, and synchronously associate the water depth, heading and attitude data through timestamps to generate an original point cloud dataset carrying spatial attitude information; Compressing and encrypting the original point cloud dataset through an edge computing node to generate the monitoring point cloud dataset; The monitoring point cloud dataset is mapped to the coordinate system of the dock BIM model by a seven-parameter coordinate transformation method, and a three-dimensional underwater siltation grid model is constructed using a grid interpolation algorithm, specifically including: The point cloud data corresponding to the blank area of the obstacle is supplemented by using the neighboring point interpolation method, and the noise signal in the sonar echo is eliminated by using the wavelet transform algorithm; The WGS-84 coordinate system of the heading data is converted to the coordinate system of the dock BIM model through the seven-parameter coordinate conversion method. The corresponding coordinate system conversion formula is as follows: Where (X, Y, Z) represents the coordinates in the coordinate system of the terminal BIM model; δS is the scale factor; ε x , ε y , ε z is the rotation parameter; (X0, Y0, Z0) represents the coordinates in the WGS-84 coordinate system; ΔX, ΔY, ΔZ represent the translation parameters; Using the grid interpolation algorithm, the three-dimensional underwater siltation grid model is generated based on the monitoring point cloud data set; Acquiring the spectral characteristics of the sonar echo and inputting them into a random forest model, automatically distinguishing natural sedimentation elements from artificial waste elements in the three-dimensional underwater sedimentation grid model, and automatically calculating the sedimentation volume, sedimentation thickness and scour boundary position of the natural sedimentation elements; A three-color heat map is generated based on the siltation thickness, and is superimposed on the three-dimensional underwater siltation grid model in real time through the API interface of the dock BIM model. A dynamic siltation change twin is generated and sent to the operation and maintenance end. The dynamic siltation change twin supports the operation and maintenance end to view the siltation change status of any area through interactive operations; The method of using the grid interpolation algorithm to generate the three-dimensional underwater siltation grid model based on the monitoring point cloud dataset specifically includes: Based on the discrete data points i and j in the monitoring point cloud data set and the corresponding water depth data h i and h j , i = 1, 2, ..., N, j = 1, 2, ..., N, N is the total number of discrete data points in the monitoring point cloud data set, and the variance function value is calculated as follows: In the formula, γ(x i ,x j ) represents the variance function value between discrete data points i and j; d represents the spatial distance between discrete data points i and j; M(d) represents the total number of discrete data point pairs with a spatial distance of d; h i 、h j Represent the water depth data of discrete data points i and j respectively; Set the coordinates of the grid nodes in the three-dimensional underwater sedimentation grid model to (x m ,y n ), Among them, m and n represent the grid index; x min 、y min Indicates the minimum coordinate value of the three-dimensional underwater sedimentation grid model; represents the grid step size; The grid interpolation algorithm is used to calculate the grid nodes (x m ,y n ) The water depth data h corresponding to mn as follows: Where λ i Represents the weight coefficient, by solving the linear equations To determine, γ(x i ,x m ) represents the variation function value between the discrete data point i and the grid node; μ represents the Lagrange multiplier; The grid node (x m ,y n ) and the water depth data h mn Perform correlation to generate three-dimensional underwater sedimentation grid data points (x m ,y n ,h mn ), based on all the three-dimensional underwater sedimentation grid data points (x m ,y n ,h mn ) construct the three-dimensional underwater siltation grid model.

2. The method for monitoring underwater siltation of a dock based on a BIM model according to claim 1, characterized in that: The underwater sedimentation monitoring area is delineated based on the coordinate system of the dock BIM model and a preset navigation monitoring route is generated, specifically including: Obtaining the terminal BIM model, parsing coordinate data in the terminal BIM model, and establishing a three-dimensional spatial reference system, i.e., a coordinate system of the terminal BIM model; The visual interface of the dock BIM model supports manual or automatic selection of the underwater sedimentation monitoring area, and automatically filters out invalid areas in the underwater sedimentation monitoring area in combination with historical water depth data; The preset navigation monitoring route is generated based on the underwater sedimentation monitoring area by using the equidistant parallel line method.

3. The method for monitoring underwater siltation of a dock based on a BIM model according to claim 2, characterized in that: The method of generating the preset navigation monitoring route based on the underwater sedimentation monitoring area by using the equidistant parallel line method specifically includes: The underwater sedimentation monitoring area is divided into a regular rectangular monitoring area and a concave-convex shoreline monitoring area. For the regular rectangular monitoring area, the preset navigation monitoring routes are distributed in parallel and back; for the concave-convex shoreline monitoring area, the preset navigation monitoring routes automatically adapt to the boundary contour of the concave-convex shoreline monitoring area; The initial survey line spacing of the preset navigation monitoring route is set, and the historical sedimentation database is called to obtain the historical sedimentation rate. For the underwater sedimentation monitoring area where the historical sedimentation rate is greater than the maximum sedimentation rate threshold, the initial survey line spacing is automatically shortened to the minimum survey line spacing threshold; for the area where the historical sedimentation rate is less than the minimum sedimentation rate threshold, the initial survey line spacing is automatically relaxed to the maximum survey line spacing threshold.

4. The method for monitoring underwater siltation of a dock based on a BIM model according to claim 1, characterized in that: Based on the three-dimensional underwater siltation grid model, the time series analysis algorithm is used to predict future siltation trends and obtain the current siltation warning level. Combined with the actual siltation cost, the optimal siltation area and engineering volume are automatically recommended to the operation and maintenance end. Specifically, the following steps are performed: Calculating a current sedimentation rate of each grid unit area in the three-dimensional underwater sedimentation grid model, and identifying an area where the current sedimentation rate is greater than a critical sedimentation rate threshold as a sedimentation warning area; For the sedimentation warning area, using a long short-term memory network model to predict the future sedimentation thickness and output the future sedimentation trend; A three-level siltation warning mechanism and a siltation thickness warning threshold for each level are set. When the future siltation thickness reaches the siltation thickness warning threshold for the corresponding level, the optimal siltation area and engineering volume are automatically recommended to the operation and maintenance end in combination with the actual siltation cost.

5. A dock underwater siltation monitoring system based on a BIM model, applied to a dock underwater siltation monitoring method based on a BIM model as claimed in any one of claims 1 to 4, characterized in that: The monitoring system comprises: The route determination module is used to define the underwater sedimentation monitoring area based on the coordinate system of the dock BIM model and generate a preset navigation monitoring route; A data acquisition module is used for the unmanned vessel to travel according to the preset navigation monitoring route and synchronously collect water depth, heading and attitude data through a single-beam bathymetry system to form a monitoring point cloud data set; A model construction module is used to map the monitoring point cloud dataset to the coordinate system of the dock BIM model through a seven-parameter coordinate transformation method, and to construct a three-dimensional underwater siltation grid model using a grid interpolation algorithm; The result recommendation module is used to predict the future sedimentation trend and obtain the current sedimentation warning level based on the three-dimensional underwater sedimentation grid model using a time series analysis algorithm, and automatically recommend the optimal sedimentation area and project volume to the operation and maintenance end in combination with the actual sedimentation cost.

6. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor runs the computer program stored in the memory, the processor executes the steps of the method for monitoring underwater siltation of a dock based on a BIM model as described in any one of claims 1 to 4.

7. A readable storage medium storing a computer program, wherein: When the computer program is executed by a processor, it is used to implement the steps of a method for monitoring underwater siltation of a dock based on a BIM model as described in any one of claims 1 to 4.

Images