Trailing suction dredger broadside adaptive dynamic enhancement method and system based on ice load prediction

Abstract

The invention relates to a trailing suction dredger broadside adaptive dynamic enhancement method and system based on ice load prediction. The method comprises the following steps that S1, ice layer space distribution and floating ice physical parameters are obtained through multi-source sensor fusion; wherein the physical parameters of the floating ice comprise the thickness, the shape and the movement speed of the floating ice; s2, establishing a three-dimensional ice rink dynamic model with a weight; wherein the weight comprises the thickness of the floating ice, the movement speed of the floating ice and the collision angle of the floating ice colliding with the broadside of the ship; s3, predicting the ice load grade of each region of the ship broadside based on an LSTM neural network prediction model, and obtaining a prediction result; wherein the broadside of the ship is partitioned, and at least one protection module is arranged in each partition; and S4, dynamically adjusting the damping coefficient of the protection module according to the prediction result. Through multi-source data fusion, dynamic adjustment and modular replacement technologies, self-adaptive protection of the broadside structure in ice region operation is achieved.

Description

Technical Field

[0001] This invention relates to the interdisciplinary field of marine engineering and intelligent control, and in particular to an adaptive dynamic enhancement method and system for the hull side of a trailing suction hopper dredger based on ice load prediction. Background Technology

[0002] Traditional trailing suction hopper dredgers face severe challenges operating in ice-covered areas. Existing ice-strengthening solutions mainly employ static structural reinforcement techniques, relying on passive defense measures such as simply increasing steel plate thickness and densely arranging reinforcing ribs to cope with the polar environment. This extensive reinforcement approach has gradually revealed systemic defects in engineering practice: for example, increased structural weight, decreased speed at full load, and increased fuel consumption.

[0003] More importantly, static structural reinforcement schemes cannot effectively respond to the complex dynamic ice loads in the Arctic Ocean. When navigating in ice-covered areas, parameters such as ice thickness, density, and speed of movement exhibit dynamic changes on a minute-by-minute basis, and traditional rigid structures cannot achieve adaptive load adjustment. Therefore, adaptive dynamic reinforcement methods and systems for trailing suction hopper dredgers based on ice load prediction are becoming a core research direction for polar engineering vessels. Summary of the Invention

[0004] To address the problems existing in the prior art, this invention provides a method and system for adaptive dynamic reinforcement of the hull side of a trailing suction hopper dredger based on ice load prediction. This method achieves adaptive protection of the hull side structure during ice-covered operations through multi-source data fusion, dynamic adjustment, and modular replacement technology, solving the problems of weight redundancy, poor dynamic adaptability to ice loads, and low maintenance efficiency caused by static structural reinforcement.

[0005] This invention is implemented as follows: an adaptive dynamic enhancement method for the hull side of a trailing suction hopper dredger based on ice load prediction, comprising the following steps: S1. Obtain the spatial distribution of ice layers and physical parameters of floating ice through multi-source sensor fusion; Among them, the physical parameters of the floating ice include its thickness, shape, and speed of movement; S2. Establish a weighted three-dimensional dynamic model of the ice rink; The weights include ice thickness, ice speed, and the impact angle of the ice hitting the ship's side; S3. Predict the ice load level in each region of the ship's side based on the LSTM neural network prediction model, and obtain the prediction results; The ship's side compartments include at least one protective module in each compartment. S4. Dynamically adjust the damping coefficient of the protection module based on the prediction results.

[0006] In the above technical solution, preferably, the specific process of obtaining the spatial distribution of the ice layer and the physical parameters of the floating ice in step S1 is as follows: The shipborne lidar scans a fan-shaped area within a certain distance in front of the ship, generates point cloud data, and determines the thickness and shape of the floating ice; Millimeter-wave radar monitors the speed of ice floes within a certain range in real time; The satellite ice image receiving module acquires raster data of ice density synthesized by polar-orbiting satellites; In the above technical solution, a further preferred embodiment is the following process for establishing a weighted three-dimensional dynamic model of the ice rink in step S2: An Euler-Lagrange coordinate system was established, and the lidar point cloud data was registered to the ship's relative coordinate system through an iterative nearest point algorithm. The ice density raster data received from satellite ice images was then fused to generate a weighted three-dimensional ice field model.

[0007] In the above technical solution, preferably, the specific process of predicting the ice load level of each region on the ship's side in step S3 is as follows: The embedded AI processor runs an LSTM neural network prediction model; Preprocessing: Cleaning and standardizing the raw data, and unifying the units of measurement; Feature construction: The input features of the LSTM neural network prediction model include: ice thickness, ice movement speed, ice density, and ice contact angle. First, the original data of the input features are fused with multi-source data to construct high-value features, including: effective ice thickness, relative velocity vector between ship speed and ice movement speed, expected contact time between ship and ice, ice contact angle, etc. Feature extraction: short-term patterns in the first layer learning sequence and long-term patterns in the second layer learning sequence of the LSTM neural network prediction model; Attention mechanism: The LSTM neural network prediction model automatically calculates which weight information at which moments in the past period is most critical to predicting the next impact.

[0008] Decision: The fully connected layer maps the final features extracted by the LSTM neural network prediction model into multiple output neurons, each representing a different level of probability.

[0009] Output: Select the level with the highest probability as the prediction result and send it to the decision control system.

[0010] In the above technical solution, preferably, the protection module includes a variable valve, an oil chamber, and a magnetic attraction device. The magnetic attraction device is part of the protection module and is magnetically attached to the outer side plate of the ship. The oil chamber is installed on the magnetic attraction device, and the interior of the oil chamber is filled with hydraulic oil. The oil chamber is divided into two parts and connected by a variable valve.

[0011] In the above technical solution, preferably, the specific process of dynamically adjusting the damping coefficient of the protection module in step S4 is as follows: By changing the size of the variable valve in the side protection module to switch the damping coefficient, the corresponding levels are 1, 2, 3, and 4 for impact forces ranging from 0-200kN, 201-400kN, 401-600kN, and >600kN, respectively, and the corresponding damping coefficients are standby mode, low damping mode, medium damping mode, and high damping mode, respectively.

[0012] In the above technical solution, preferably, the following steps are also included: S5. Damage Detection: At least one strain gauge is arranged on the inner side of the outer surface of each protective module. The strain gauges of all protective modules form a strain gauge array, and the equivalent plastic strain of the protective module is monitored based on the strain gauge array. pl, when the equivalent plastic strain If the damage threshold is reached, the corresponding protection module will be quickly disassembled and replaced.

[0013] In the above technical solution, a further preferred embodiment is the damage threshold: pl≥0.2 for 2 seconds, or single impact strain peak ≥ 0.5.

[0014] In the above technical solution, a further preferred embodiment is that the process of quickly disassembling and replacing the protective module is as follows: Electromagnetic latch releases upon power failure: Cut off the electromagnetic latch current, and the spring pin will pop out the protection module, allowing you to remove the damaged protection module. Lifting and positioning: The drone hoists the new protective module to the target area, uses ultra-wideband positioning for coarse alignment, laser scanning to match the edge features of the protective module, and electromagnetic latches are energized to attract it with an attraction force ≥1800N. System self-check: The hydraulic pipeline pressure was restored to 5MPa±0.5MPa, the strain gauge error was <0.05%, and the protection module replacement was completed.

[0015] In the above technical solution, preferably, the surface of the ship's side outer plating is coated with an ice-repellent coating, the contact angle of which is ≥150°, for use in ice-affected areas.

[0016] An adaptive dynamic enhancement system for the hull side of a trailing suction hopper dredger based on ice load prediction includes: Sensing layer: used to acquire the spatial distribution of ice layers and the physical parameters of floating ice; Includes shipborne lidar (LiDA), millimeter-wave radar, and satellite ice image receiving module; The shipborne lidar LiDA is used to scan a fan-shaped area 0-150m in front of the ship, generate point cloud data, and determine the thickness of the floating ice. Millimeter-wave radar is used to monitor the speed of ice floes within a 200-meter range in real time; The satellite ice image receiving module is used to acquire sea ice concentration raster data synthesized by polar-orbiting satellites; Decision-making level: Used to predict ice load levels in various areas along the ship's side; This includes an LSTM neural network prediction model and an embedded AI processor; the LSTM neural network prediction model is built into the embedded AI processor. Execution layer: Used to dynamically adjust the damping coefficient of the protection module based on the prediction results; It includes protective modules, which are placed on the sides of the ship.

[0017] Compared with the prior art, the advantages and positive effects of this invention are: This invention predicts the impact intensity of floating ice through multi-source data fusion and, combined with different load levels, achieves adaptive stiffness adjustment and rapid modular replacement of the side protection module structure. It breaks through the traditional static strengthening mode, significantly improving the economy and safety of polar operation vessels. For the first time, it integrates ice load prediction, structural response, and modular replacement to achieve closed-loop control of "perception-decision-execution."

[0018] Impact resistance: The ice load resistance of the side structure is improved through real-time prediction and dynamic adjustment.

[0019] Economic benefits: Modular replacement reduces downtime losses and lowers maintenance costs.

[0020] Environmental friendliness: It does not increase the weight of the ship, reduces the load on the diesel engine, and reduces carbon emissions. Attached Figure Description

[0021] Figure 1 This is a schematic diagram of ice load prediction and side protection provided in an embodiment of the present invention; Figure 2 This is a cross-sectional view of the protection module provided in an embodiment of the present invention; Figure 3 This is a flowchart illustrating the specific process of predicting the ice load level in various regions of a ship's side, as provided in this embodiment of the invention. Figure 4 This is a system architecture diagram provided in an embodiment of the present invention.

[0022] In the diagram: 1. Protection module; 2. Radar scanning; 3. Variable valve; 4. Oil chamber; 5. Magnetic suction device. Detailed Implementation

[0023] To further understand the invention's content, features, and effects, the following embodiments are provided, and detailed descriptions are given below in conjunction with the accompanying drawings: In the description of this invention, it should be understood that the terms "upper", "lower", "front", "rear", "left", "right", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0024] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0025] Example Please see Figures 1-3 The present invention provides an adaptive dynamic enhancement method for the hull side of a trailing suction hopper dredger based on ice load prediction, comprising the following steps: S1. Obtain the spatial distribution of ice layers and physical parameters of floating ice through multi-source sensor fusion; The physical parameters of ice floes include their thickness, shape, and speed of movement.

[0026] S2. Establish a weighted three-dimensional dynamic model of the ice rink; The weights include ice thickness, ice speed, and the impact angle of the ice hitting the ship's side; determining the posture of the ice hitting the ship's side is helpful for later assessment of the magnitude of its impact force on the side.

[0027] S3. Based on the LSTM (Long Short-Term Memory) neural network prediction model, predict the ice load level in each area of ​​the ship's side and obtain the prediction results; The ship's side is divided into zones, each zone containing at least one protective module 1; several protective modules are arranged on the ship's side, and the ship's side is divided into zones (each 2m×2m is a zone) for each protective module, so that each zone contains at least one protective module.

[0028] S4. Dynamically adjust the damping coefficient of the protection module based on the prediction results.

[0029] As a preferred embodiment, the specific process for obtaining the spatial distribution of the ice layer and the physical parameters of the floating ice in step S1 is as follows: Shipborne LiDAR emits a 905nm laser beam at a frequency of 20Hz to scan a fan-shaped area (e.g., 0-150m) in front of the ship, generating point cloud data to determine the thickness and shape of the floating ice (accuracy ±5cm). The millimeter-wave radar uses a 77GHz frequency modulated continuous wave (FMCW) mode to monitor the speed of ice floes within a certain range (e.g., 200m) in real time (resolution 0.1m / s). The satellite ice image receiving module acquires ice density raster data (500m×500m resolution) synthesized by polar-orbiting satellites every 5 minutes through the BeiDou short message channel.

[0030] As a preferred implementation, the specific process of establishing a weighted three-dimensional dynamic model of the ice rink in step S2 is as follows: An Euler-Lagrange coordinate system was established, and the LiDAR point cloud data was registered to the ship's relative coordinate system using the Iterative Closest Point (ICP) algorithm. The ice density raster data received from satellite ice images was then fused to generate a weighted three-dimensional ice field model.

[0031] As a preferred embodiment, the specific process for predicting the ice load level in each region of the ship's side in step S3 is as follows: The embedded AI processor (20 cores) runs an LSTM neural network prediction model.

[0032] Preprocessing: Cleaning and standardizing the raw data, and unifying the units of measurement; Data cleaning primarily refers to removing outliers, errors, and missing values ​​from raw data to ensure data quality and consistency. Data standardization refers to scaling data with different characteristics to the same numerical scale. This is because the optimization algorithms of neural network prediction models such as LSTM are highly sensitive to the scale of the input.

[0033] Feature Construction: The input features of the LSTM neural network prediction model include: ice thickness, ice movement speed, ice density, and ice contact angle. The raw data of the input features cannot be directly input. The raw data of the input features need to be fused with multi-source data to construct high-value features, including: effective ice thickness, relative velocity vector between ship speed and ice movement speed, expected contact time between ship and ice, and ice contact angle.

[0034] Feature extraction: The LSTM neural network prediction model learns short-term patterns in the first layer of the learning sequence and long-term patterns in the second layer. Specifically, short-term patterns refer to the transient state of the ice floes, such as the trend of their movement speed, which is equivalent to adopting a posture to prepare for a collision when encountering an ice floe. Long-term patterns refer to longer-term changes, such as the overall interaction between the boat and the ice rink. After long-term learning, the computational burden on the system can be reduced for the attitude changes of the entire ice rink, which is more advantageous than short-term patterns.

[0035] Attention mechanism: The LSTM neural network prediction model automatically calculates which time-related information is most critical to predicting the next impact over a period of time (for example, the information in the most recent 2 seconds may have the highest weight).

[0036] Decision: The fully connected layer maps the final features extracted by the LSTM neural network prediction model into multiple output neurons, each representing a different level of probability. Specifically, it is set to have 4 output neurons, representing probabilities of levels 1-4, meaning that the LSTM neural network prediction model outputs the impact level of the side partition as levels 1-4.

[0037] Output: Select the level with the highest probability as the prediction result and send it to the decision control system.

[0038] The LSTM neural network prediction model for predicting ice load on the side of a trailing suction hopper dredger in this invention is a model that has been initially trained using a large-scale general dataset, and its weights already include learned general features and temporal patterns.

[0039] In a preferred embodiment, the protective module 1 includes a variable valve 3, an oil chamber 4, and a magnetic attraction device 5. The magnetic attraction device 5 is part of the protective module 1 and is magnetically attached to the outer side plating of the ship. The oil chamber 4 is mounted on the magnetic attraction device 5 and contains hydraulic oil. The oil chamber 4 is divided into two parts and connected by the variable valve 3. When the protective module 1 is subjected to an external impact, the outer oil chamber is squeezed towards the inner oil chamber, thus delaying the impact.

[0040] In a preferred embodiment, the specific process of dynamically adjusting the damping coefficient of the protection module in step S4 is as follows: By changing the size of the variable valve in the side protection module to switch the damping coefficient, the corresponding levels are 1, 2, 3, and 4 for impact forces ranging from 0-200kN, 201-400kN, 401-600kN, and >600kN, respectively, and the corresponding damping coefficients are standby mode, low damping mode, medium damping mode, and high damping mode, respectively.

[0041] The grading thresholds are shown in Table 1: Table 1. Ice Layer Classification Threshold Table

[0042] As a preferred embodiment, the adaptive dynamic enhancement method for the hull side of a trailing suction hopper dredger based on ice load prediction further includes the following steps: S5. Damage Detection: At least one strain gauge is arranged on the inner side of the outer surface of each protective module (the strain gauge density is 4 gauges / m). 2 The strain gauges of each protection module form a strain gauge array, and the equivalent plastic strain of the protection module is monitored based on the strain gauge array. pl, when the equivalent plastic strain If the damage threshold is reached, the corresponding protection module will be quickly disassembled and replaced.

[0043] Wherein, the damage threshold is: pl≥0.2 for 2 seconds, or single impact strain peak ≥ 0.5.

[0044] The process for quick disassembly and replacement of the protective module is as follows: Electromagnetic latch releases upon power failure: Cut off the electromagnetic latch current (rated holding force 1500N), and the spring pin (stiffness coefficient 30N / mm) will pop out the protective module, allowing you to remove the damaged protective module. Lifting and positioning: The drone hoisted the new protective module to the target area, and used ultra-wideband (UWB) positioning for coarse alignment (error ±50mm). Laser scanning matched the edge features of the protective module (accuracy ±1mm). The electromagnetic latch was powered on and attracted, with an attraction force ≥1800N. System self-check: The hydraulic pipeline pressure was restored to 5MPa±0.5MPa, the strain gauge error was <0.05%, and the protection module replacement was completed.

[0045] The protective module can be replaced via a magnetic quick-release interface with a tensile strength of ≥200MPa and a self-alignment function.

[0046] The damping coefficient of the oil chamber of the protection module can be adjusted from 10 to 50 kN·s / m, and the response time is <0.5 seconds.

[0047] As a preferred embodiment, the surface of the ship's side outer plating is coated with an ice-repellent coating with a contact angle ≥150° for ice protection.

[0048] Please see Figure 4 An adaptive dynamic enhancement system for the hull side of a trailing suction hopper dredger based on ice load prediction, comprising: Sensing layer: used to acquire the spatial distribution of ice layers and the physical parameters of floating ice; Includes shipborne lidar (LiDA), millimeter-wave radar, and satellite ice image receiving module; The shipborne lidar LiDA is used to scan a fan-shaped area 0-150m in front of the ship, generate point cloud data, and determine the thickness of the floating ice. Millimeter-wave radar is used to monitor the speed of ice floes within a 200-meter range in real time; The satellite ice image receiving module is used to acquire sea ice concentration raster data synthesized by polar-orbiting satellites; Decision-making level: Used to predict ice load levels in various areas along the ship's side; This includes an LSTM neural network prediction model and an embedded AI processor; the LSTM neural network prediction model is built into the embedded AI processor. Execution layer: Used to dynamically adjust the damping coefficient of the protection module based on the prediction results; It includes protective modules, which are placed on the sides of the ship.

[0049] To better understand the above embodiments of the present invention, further explanation is provided below with reference to specific examples.

[0050] Example 1 A certain 25000m 3 Trailing hopper dredgers operating in the Arctic shipping route Data collection: The LiDAR scanner scanned the ice layer 150m ahead and detected a group of ice floes with an average thickness of 1.2m. Satellite data updates show that the ice layer is moving at a speed of 0.8 m / s.

[0051] Forecasting and Decision Making: The LSTM neural network predicts that the mid-starboard region will experience a level 3 impact (predicted force 520kN) in 6 seconds. The decision control system activates the "shock resistance mode" of the target area.

[0052] Execution response: The damping coefficient of the protection module has been adjusted to medium damping mode.

[0053] Example 2 Protection module replacement operation The damaged protection module automatically shuts off power and releases the electromagnetic latch; The drone was used to hoist the new protective module, and laser guidance was used to complete the positioning. The magnetic force is restored after the system self-test passes.

[0054] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Any simple modifications, equivalent changes, and alterations made to the above embodiments based on the technical essence of the present invention shall fall within the scope of the technical solution of the present invention.

Claims

1. A method for adaptive dynamic enhancement of the hull side of a trailing suction hopper dredger based on ice load prediction, characterized in that, Includes the following steps: S1. Obtain the spatial distribution of ice layers and physical parameters of floating ice through multi-source sensor fusion; Among them, the physical parameters of the floating ice include its thickness, shape, and speed of movement; S2. Establish a weighted three-dimensional dynamic model of the ice rink; The weights include ice thickness, ice speed, and the impact angle of the ice hitting the ship's side; S3. Predict the ice load level in each region of the ship's side based on the LSTM neural network prediction model, and obtain the prediction results; The ship's side compartments include at least one protective module in each compartment. S4. Dynamically adjust the damping coefficient of the protection module based on the prediction results.

2. The adaptive dynamic enhancement method for the hull side of a trailing suction hopper dredger based on ice load prediction as described in claim 1, characterized in that, In step S1, the specific process of obtaining the spatial distribution of the ice layer and the physical parameters of the floating ice is as follows: The shipborne lidar scans a fan-shaped area within a certain distance in front of the ship, generates point cloud data, and determines the thickness and shape of the floating ice; Millimeter-wave radar monitors the speed of ice floes within a certain range in real time; The satellite ice image receiving module acquires raster data of ice density synthesized by polar-orbiting satellites.

3. The adaptive dynamic enhancement method for the hull side of a trailing suction hopper dredger based on ice load prediction as described in claim 2, characterized in that, In step S2, the specific process of establishing a weighted three-dimensional dynamic model of the ice rink is as follows: An Euler-Lagrange coordinate system was established, and the lidar point cloud data was registered to the ship's relative coordinate system through an iterative nearest point algorithm. The ice density raster data received from satellite ice images was then fused to generate a weighted three-dimensional ice field model.

4. The adaptive dynamic enhancement method for the hull side of a trailing suction hopper dredger based on ice load prediction as described in claim 1, characterized in that, In step S3, the specific process for predicting the ice load level in each region of the ship's side is as follows: The embedded AI processor runs an LSTM neural network prediction model; Preprocessing: Cleaning and standardizing the raw data, and unifying the units of measurement; Feature construction: The input features of the LSTM neural network prediction model include: ice thickness, ice movement speed, ice density, and ice contact angle. First, the original data of the input features are fused with multi-source data to construct high-value features, including: effective ice thickness, relative velocity vector between ship speed and ice movement speed, expected contact time between ship and ice, ice contact angle, etc. Feature extraction: short-term patterns in the first layer learning sequence and long-term patterns in the second layer learning sequence of the LSTM neural network prediction model; Attention mechanism: The LSTM neural network prediction model automatically calculates which weight information at which moments in the past period is most critical to predicting the impact of the next event. Decision: The fully connected layer maps the final features extracted by the LSTM neural network prediction model into multiple output neurons, each representing a different level of probability; Output: Select the level with the highest probability as the prediction result and send it to the decision control system.

5. The adaptive dynamic enhancement method for the hull side of a trailing suction hopper dredger based on ice load prediction according to claim 1, characterized in that, The protective module includes a variable valve, an oil chamber, and a magnetic suction device. The magnetic suction device is part of the protective module and is magnetically attached to the outer side plating of the ship. The oil chamber is installed on the magnetic suction device and contains hydraulic oil. The oil chamber is divided into two parts and connected by the variable valve.

6. The adaptive dynamic enhancement method for the hull side of a trailing suction hopper dredger based on ice load prediction as described in claim 1, characterized in that, In step S4, the specific process of dynamically adjusting the damping coefficient of the protection module is as follows: By changing the size of the variable valve in the side protection module to switch the damping coefficient, the corresponding levels are 1, 2, 3, and 4 for impact forces ranging from 0-200kN, 201-400kN, 401-600kN, and >600kN, respectively, and the corresponding damping coefficients are standby mode, low damping mode, medium damping mode, and high damping mode, respectively.

7. The adaptive dynamic enhancement method for the hull side of a trailing suction hopper dredger based on ice load prediction according to claim 1, characterized in that, It also includes the following steps: S5. Damage Detection: At least one strain gauge is arranged on the inner side of the outer surface of each protective module. The strain gauges of all protective modules form a strain gauge array, and the equivalent plastic strain of the protective module is monitored based on the strain gauge array. pl, when the equivalent plastic strain If the damage threshold is reached, the corresponding protection module will be quickly disassembled and replaced.

8. The adaptive dynamic enhancement method for the hull side of a trailing suction hopper dredger based on ice load prediction according to claim 7, characterized in that, The damage threshold: pl≥0.2 for 2 seconds, or single impact strain peak ≥ 0.5; The process for quick disassembly and replacement of the protective module is as follows: Electromagnetic latch releases upon power failure: Cut off the electromagnetic latch current, and the spring pin will pop out the protection module, allowing you to remove the damaged protection module. Lifting and positioning: The drone hoists the new protective module to the target area, uses ultra-wideband positioning for coarse alignment, laser scanning to match the edge features of the protective module, and electromagnetic latches are energized to attract it with an attraction force ≥1800N. System self-check: The hydraulic pipeline pressure was restored to 5MPa±0.5MPa, the strain gauge error was <0.05%, and the protection module replacement was completed.

9. The adaptive dynamic enhancement method for the hull side of a trailing suction hopper dredger based on ice load prediction according to claim 1, characterized in that, The ship's side outer plating is coated with an ice-repellent coating with a contact angle ≥150° for ice protection.

10. An adaptive dynamic enhancement system for the hull of a trailing suction hopper dredger based on ice load prediction, the system being used to implement the adaptive dynamic enhancement method according to any one of claims 1 to 9, characterized in that, include: Sensing layer: used to acquire the spatial distribution of ice layers and the physical parameters of floating ice; Includes shipborne lidar (LiDA), millimeter-wave radar, and satellite ice image receiving module; The shipborne lidar LiDA is used to scan a fan-shaped area 0-150m in front of the ship, generate point cloud data, and determine the thickness of the floating ice. Millimeter-wave radar is used to monitor the speed of ice floes within a 200-meter range in real time; The satellite ice image receiving module is used to acquire sea ice concentration raster data synthesized by polar-orbiting satellites; Decision-making level: Used to predict ice load levels in various areas along the ship's side; This includes an LSTM neural network prediction model and an embedded AI processor; the LSTM neural network prediction model is built into the embedded AI processor. Execution layer: Used to dynamically adjust the damping coefficient of the protection module based on the prediction results; It includes protective modules, which are placed on the sides of the ship.

Images