A method for optimizing the height automatic adjustment of the energy dissipation device of a trailing suction hopper dredger

Abstract

The present invention provides a method for optimizing the automatic adjustment of the height of the energy dissipation device of a trailing suction hopper dredger, comprising four steps: determining core parameters, simulating the loading process, analyzing the loading efficiency, and outputting optimized height of the energy dissipation device; establishing a prediction model through numerical simulation calculation and convolutional neural network to obtain a recommended value for the height of the energy dissipation device. One of the systems for optimizing the height of the energy dissipation device on a trailing suction hopper dredger comprises a core parameter input module, a loading process simulation prediction module, a loading efficiency analysis module, a design optimization output module, and an automatic height adjustment module. The present invention can achieve automatic and flexible adjustment of the height of the energy dissipation device, and improve the loading efficiency and the smoothness of the sediment loading in the cabin by reducing overflow losses.

Description

Technical Field

[0001] The invention belongs to the technical field of dredging engineering, and in particular relates to an automatic adjustment and loading method of an energy dissipation system of a trailing suction dredger. Technical Background

[0002] Trailing suction hopper dredgers are the workhorse vessels of the dredging industry, playing a vital role in projects such as channel dredging, land reclamation, and port maintenance. The mud tank system is a crucial component of a trailing suction hopper dredger, responsible for dredging, loading, and flushing the tank to the shore. The loading process of a trailing suction hopper dredger can be divided into two parts: loading and overflow, depending on the characteristics of each stage. Based on the target soil properties, mud (a mixture of water and sediment) continuously flows into the mud tank through the tank inlet pipe. Part of the sediment settles to the bottom of the tank under the action of gravity, while the other part of the sediment mixes with water and remains suspended in the tank. When the liquid level of this water-sand mixture exceeds the height of the overflow tube, varying degrees of overflow loss will occur.

[0003] The efficiency of the loading and overflow construction of a trailing suction hopper dredger is directly related to the design and layout of the energy dissipation device. At present, the energy dissipation devices of most domestic trailing suction hopper dredgers are of fixed design. They cannot adjust the height of the energy dissipation device to various complex working conditions such as different soil types and mud concentrations, and cannot specifically reduce the disturbance of the sediment in the cabin caused by the incoming water flow. When the sediment particles are coarse, the sediment will settle quickly, and if the energy dissipation device is installed too low, it will easily block the outflow port. When the sediment particles are fine, if the energy dissipation device is installed too high, the fine sediment will be disturbed by the incoming water flow, causing most of it to be suspended, increasing overflow losses, reducing loading efficiency, and at the same time having an adverse impact on the surrounding construction water environment. Summary of the Invention

[0004] In response to the shortcomings of the existing technology, the present invention provides an energy dissipation device of a trailing suction hopper dredger with adjustable height, an automatic height adjustment optimization method and a ship-borne system, which can realize automatic and flexible adjustment of the height of the energy dissipation device during the construction of the trailing suction hopper dredger, and improve the loading flatness and loading efficiency of the sediment in the cabin by minimizing the disturbance of the sediment in the cabin and reducing overflow losses. It can meet the needs of various working conditions such as different soil types, inlet concentrations, inlet flow rates, and the structural dimensions of the mud cabin of the trailing suction hopper dredger.

[0005] Technical Solution

[0006] One of the purposes of the present invention is to provide a method for optimizing the automatic height adjustment of an energy dissipation device of a trailing suction hopper dredger, comprising:

[0007] Process 1: Constructing an adjustable height energy dissipation device in the cabin of the suction dredger;

[0008] Process 2 determines the optimal energy dissipation height;

[0009] Step 1: Determine core parameters using existing trailing suction hopper construction data, geological survey reports of the construction area, and soil condition data. Core parameters include: geometric model, boundary conditions, grid conditions, initial conditions, fluid motion, and sediment motion. Specific parameters are as follows:

[0010] (1) The geometric model mainly includes: the dimensions of the mud tank structure, the dimensions of the transverse and longitudinal triangular tanks, the dimensions of the overflow equipment, the dimensions and height of the energy dissipation equipment, and the positions of the transverse and longitudinal triangular tanks and the energy dissipation overflow equipment relative to the mud tank;

[0011] (2) Boundary conditions include: the inlet boundary is the loading flow rate and loading mud concentration, the overflow boundary is the free outflow boundary, the upper boundary of the mud tank is the atmospheric pressure boundary, and the bulkhead boundary is the wall boundary with a certain roughness material;

[0012] (3) Grid conditions include: grid range in the x, y, and z directions and unit grid size.

[0013] (4) Initial conditions include: initial water level H0, loading simulation time T;

[0014] (5) Fluid motion parameters include: fluid temperature T, fluid viscosity μ, and fluid motion equation.

[0015] (6) The fluid motion equation uses the RNG k-ε equation according to the characteristics of the loading compartment, as follows:

[0016]

[0017]

[0018] Where: k——turbulent kinetic energy;

[0019] u i - velocity component;

[0020] x i 、y i - coordinate components;

[0021] μ——molecular viscosity coefficient;

[0022] μ t ——turbulent viscosity coefficient;

[0023] G k — turbulent energy generation term;

[0024] ε——turbulent energy dissipation rate;

[0025] α k , α ε ——k, ε are the turbulent Prandtl numbers;

[0026] Among them C 1ε =1.42, η0=4.377, β=0.012;

[0027] in C 2ε =1.68.

[0028] Sediment movement parameters primarily include particle size, density, porosity, critical Shields number, underwater angle of repose, propulsion coefficient, and scour coefficient. These parameters encompass four types of movement: suspension, deposition, scour, and propulsion. The following empirical formula is used to automatically calculate the movement pattern of each sediment particle during the simulation.

[0029] Among them, the empirical formula of Mastbergen and Von den Berg is used to calculate the sediment initiation and scouring process.

[0030]

[0031]

[0032] Where α refers to the entrainment coefficient, n s is the outer normal direction of the bed surface, θ refers to the Shields number of the sediment particles, θ cr refers to the critical Shields number of sediment particles, g refers to the acceleration of gravity, μ is the dynamic viscosity of the fluid, and d s Refers to the sediment particle size, ρ s is the sediment density, ρ f is the fluid density.

[0033] The expression of Shields number θ of sediment particles is:

[0034]

[0035] Where τ is the shear stress on the riverbed surface.

[0036] In non-horizontal bed areas, considering the influence of the sediment repose angle, the critical Shields number correction value θ cr,i 'for

[0037]

[0038] Where κ is the angle between the vertical line of the sediment surface and the gravitational acceleration g; φ is the angle of repose of the sediment; and ψ is the angle between the upslope direction of the fluid and the flow direction.

[0039] use The empirical formula of equat i is used to calculate the bed load transport process during the simulation.

[0040]

[0041] In the formula, the calculation formula for the thickness of bed load sediment δ is:

[0042]

[0043] Therefore, the calculation formula for bed load transport velocity is:

[0044]

[0045] Step 2: Carry out simulation prediction of the loading process, including two steps: establishing a loading numerical model and loading simulation prediction.

[0046] Establishing a numerical model for loading: refers to establishing a numerical model capable of simulating the loading construction process based on the input core parameters using CFD software (existing technology) to achieve loading simulation function.

[0047] Loading simulation prediction: Using the aforementioned loading numerical model, we perform numerical simulations at different energy dissipation device heights. This simulates the entire loading process and predicts how the loading volume, earthwork volume, total overflow loss, and sediment height within the tank change over time. Multiple sets of numerical simulation results are obtained for different energy dissipation device heights. The energy dissipation device height must be set manually.

[0048] The numerical simulation results also include: t (3D grid point cloud of flow at time t); ρ t (3D grid point cloud with density at time t); v x (3D grid point cloud of velocity in x direction at time t); v y (3D grid point cloud of the y-direction velocity component at time t). The results of the numerical calculation are provided to step 3.

[0049] Step 3: Based on the numerical simulation results, the evaluation criteria are set according to the four indicators of loading capacity, earthwork volume, overflow loss, and sediment flatness to obtain the optimized recommended height of the energy dissipation device;

[0050] In multiple sets of numerical simulation results, different energy dissipation device heights correspond to different four indicators: loading capacity, earthwork volume, overflow loss, and sediment flatness. An evaluation criteria sequence is set, and the recommended value for the optimal design of the energy dissipation device height is determined based on the evaluation criteria sequence (first gradient, second gradient). The loading efficiency, represented by loading capacity, earthwork volume, and overflow loss, is usually used as the first gradient judgment standard, and the sediment flatness is used as the second gradient judgment standard. Among them, the greater the loading capacity and earthwork volume, the smaller the overflow loss, and the higher the mud tank flatness, the better the impact of the energy dissipation device height on the loading effect under this working condition.

[0051] Preferably, the optimization goal is to optimize the sediment flatness α and maximize the earthwork volume D, then the optimization model objective function is:

[0052]

[0053] max D=max ∑ΔQ i ·Δt·ρ (16)

[0054] Set constraints (xt):

[0055] X = {x1,x2,x3…,x i} (17)

[0056] Among them, Δt is the length of the time period for calculating the amount of deposited earthwork; ΔQ i is the outflow rate in the Δt time period; ρ is the outflow concentration in the Δt time period; x i Belongs to [0,H], H represents the height of the mud tank, x (it) As a parameter for calculating flatness, it represents the height of sediment deposited at sampling position i at time t, It represents the average height of sediment deposited in the longitudinal centerline section of the mud chamber.

[0057] t < T (18)

[0058] Where T represents the maximum time set initially.

[0059] Set the heights of multiple energy dissipation devices and use the numerical calculation model in step 2 to calculate q t , ρ t 、v x and v y , and then calculate the flatness of the sediment and the amount of earthwork. According to the above optimization objectives (optimal flatness and maximum earthwork), the best group is selected, and the corresponding energy dissipation equipment height is the optimal energy dissipation equipment height.

[0060] Taking into account time and computing power, and based on relevant numerical calculation and prediction experience, the recommended number of energy dissipation device height groups is 12 to 20, with an interval of 0.25 to 0.5 m between each group and a height variation range of 5 to 6 m.

[0061] Process 3: Loading at the optimal energy dissipation height

[0062] Step 4: According to the optimal energy dissipation device height obtained in step 3, adjust the hydraulic support rod of the adjustable height energy dissipation device to raise or lower the height of the energy dissipation device to the optimal height.

[0063] Furthermore, in step 3, to avoid complex numerical calculations, a neural network is introduced to predict sediment height and earthwork volume.

[0064] A sediment height prediction model is established to predict sediment height as follows:

[0065] The calculation of X is done by establishing a prediction model of X(t) values ​​and the initial and boundary conditions of the numerical simulation as follows:

[0066] Input value: q t is the three-dimensional grid point cloud of the flow at time t; ρ t is the three-dimensional grid point cloud with density at time t; v x is the three-dimensional grid point cloud of the velocity in the x direction at time t; v y It is the three-dimensional grid point cloud of the y-direction velocity at time t.

[0067] Output: X(t+Δt) represents the sediment height matrix calculated for the smoothness section at the next time Δt, which serves as the label for the sediment height prediction model. The three-dimensional positions of the energy dissipation devices are all set to 0 to characterize their spatial characteristics.

[0068] The q values ​​calculated by numerical simulation are extracted at time intervals of Δt. t , ρ t 、v x 、v y q corresponding to each time t , ρ t 、v x 、v y Convolution kernel features are extracted from the 3D grid data to generate a feature matrix. The corresponding values ​​used to calculate the sediment height in the chamber at time t+Δt are combined into a label matrix X for network training. Through these operations, numerical simulation data is used to train a model for predicting sediment height at the calculation points on the flatness calculation plane.

[0069] During the training phase, the input and label are calculated by the loading process simulation prediction model in step 2; after the training is completed, the actual q t , ρ t 、v x 、v y Predict the height of sediment at each point, and then reflect the flatness of the sediment.

[0070] The earthwork volume prediction model is established to predict the earthwork volume in the cabin as follows:

[0071] As above, dt is the three-dimensional grid point cloud of earthwork volume at time t, and d(t+Δt) represents the earthwork volume calculated at the next Δt time as the label item of the earthwork volume prediction model. t , ρ t 、v x 、v yThe earthwork volume prediction model is obtained by training a network based on the data and the three-dimensional data of the earthwork volume in the mud tank at time t+Δt.

[0072] The order of the two evaluation criteria, earthwork volume and flatness, is set according to the prediction model to obtain the recommended optimized height.

[0073] A second object of the present invention is to provide a height optimization system for an energy dissipation device on a trailing suction hopper dredger, characterized by comprising a core parameter input module, a loading process simulation and prediction module, a design optimization output module, and an automatic height adjustment module, wherein these software modules are embedded in and called by the onboard loading control system;

[0074] Among them, the core parameter input module:

[0075] The core parameters include geometric model, boundary conditions, grid conditions, initial conditions, fluid motion and sediment motion; the geometric model includes: mud tank structure size, transverse and longitudinal triangular tank size, overflow equipment size, energy dissipation equipment size and height, and the position of transverse and longitudinal triangular tanks and energy dissipation overflow equipment relative to the mud tank; the boundary conditions include the inlet boundary being the loading flow and loading mud concentration, the overflow boundary being the free outflow boundary, the upper boundary of the mud tank being the atmospheric pressure boundary, and the bulkhead boundary being the wall boundary with a certain roughness material; the grid conditions include: grid range in the x, y, and z directions and unit grid size; the initial conditions include the initial water level H0 and the loading simulation time t; the fluid motion parameters include fluid temperature T, fluid viscosity μ, and fluid motion equation; the sediment motion parameters mainly include sediment particle size, sediment density, sediment porosity, critical Shields number, underwater angle of repose, push motion coefficient and scour coefficient, involving four forms of motion: sediment suspension, sediment deposition, sediment scour and sediment push;

[0076] Among them, the loading process simulation prediction module:

[0077] The core parameter input module uses the various parameters input to perform numerical calculations and simulate the loading process under different working conditions. The loading process simulation prediction module implements loading process simulation prediction, including the establishment of a loading numerical model and loading simulation prediction.

[0078] Establish a numerical model for loading: Based on the input core parameters, use CFD software (existing technology) to establish a numerical model that can simulate the loading construction process to realize the loading simulation function.

[0079] Loading simulation prediction: Using the aforementioned loading numerical model, we perform numerical simulations at different energy dissipation device heights. This simulates the entire loading process and predicts how the loading volume, earthwork volume, total overflow loss, and sediment height within the tank change over time. Multiple sets of numerical simulation results are obtained for different energy dissipation device heights. The energy dissipation device height must be set manually.

[0080] The numerical simulation results also include: t (3D grid point cloud of flow at time t); ρ t (3D grid point cloud with density at time t); v x (3D grid point cloud of velocity in x direction at time t); v y (3D grid point cloud of the y-direction velocity component at time t). The results of the numerical calculation are provided to the design optimization output module.

[0081] This module displays various calculation data in real time and visualizes simulation results within a 3D model. By viewing outputs such as specific flow rates, concentrations, and earthwork volumes during loading under various conditions, users gain a more intuitive and in-depth understanding of the loading process. This helps users analyze various loading phenomena, such as flow rate distribution and sediment concentration variations, and optimize loading plans.

[0082] Among them, the design optimization output module:

[0083] Based on the numerical simulation results output by the loading process simulation prediction module, the evaluation criteria are set in order using four indicators, including loading capacity, earthwork volume, overflow loss, and sediment flatness, to obtain the optimized recommended height of the energy dissipation device; users can click to select the order of evaluation criteria (loading efficiency, flatness) according to actual needs; the system will automatically output the optimized recommended value for the height of the energy dissipation equipment based on the evaluation criteria selected by the user.

[0084] Specifically, in multiple sets of numerical simulation results, different energy dissipation device heights correspond to different four indicators: mud tank loading capacity, earthwork volume, overflow loss, and sediment flatness. An evaluation basis sequence is set, and based on the evaluation basis sequence (first gradient, second gradient), a recommended value for the optimal design of the energy dissipation device height is generated. Usually, the loading efficiency represented by loading capacity, earthwork volume, and overflow loss is used as the first gradient judgment standard, and the sediment flatness is used as the second gradient judgment standard. Among them, the larger the loading capacity and earthwork volume, the smaller the overflow loss, and the higher the mud tank flatness, indicating that the energy dissipation device height of this working condition has a better impact on the loading effect.

[0085] Among them, the height automatic adjustment module:

[0086] Based on the recommended value for the optimized height of the energy dissipation device output by the design optimization output module, a control signal is sent via the shipboard loading control system to automatically control the hydraulic struts on the adjustable height energy dissipation device. By precisely controlling the extension and retraction of the hydraulic struts, the height of the energy dissipation device can be adjusted automatically and precisely. In addition, users can also manually determine the soil conditions of the construction area through existing geological survey data and input the appropriate height of the energy dissipation device through the loading control system based on engineering experience. After receiving the command, the system will send a control signal to the hydraulic struts to achieve manual auxiliary lifting and lowering adjustment of the energy dissipation device. This adjustment method is more flexible and can adapt to different construction environments and needs.

[0087] The beneficial effect of the present invention is to provide a method and system for automatically adjusting the height of the energy dissipation device of a trailing suction hopper dredger. During the construction process, the height of the energy dissipation device can be accurately and automatically adjusted in real time according to different soil qualities and different mud concentration requirements. The energy dissipation device is neither too low, causing rapidly settling coarse sediment to block the outlet of the energy dissipation device, nor too high, causing the incoming water flow to disturb the sediment and increase the overflow loss of suspended sediment. By accurately controlling the height of the energy dissipation device, the construction efficiency of the trailing suction hopper dredger is improved, and at the same time, the adverse effects of loading overflow on the construction water environment can be reduced, which has engineering application value. BRIEF DESCRIPTION OF THE DRAWINGS

[0088] Figure 1 This is a front view of an adjustable height energy dissipation device for a trailing suction hopper dredger according to an embodiment;

[0089] Figure 2 This is a side view of an adjustable height energy dissipation device for a trailing suction hopper dredger according to an embodiment;

[0090] Figure 3 This is a schematic diagram of the telescopic pipeline structure of an embodiment;

[0091] Figure 4 This is a schematic diagram of the structure of the hydraulic support rod for lifting in the embodiment;

[0092] Figure 5 This is a three-dimensional modeling diagram of the mud tank of the trailing suction hopper dredger in the embodiment;

[0093] Figure 6 This is the grid division of the three-dimensional model of the mud tank of the trailing suction hopper dredger in the embodiment;

[0094] Figure 7 This is a three-dimensional numerical simulation diagram of the mud tank of a trailing suction hopper dredger in the embodiment;

[0095] Figure 8 This is a flow chart of the sediment height prediction model of the embodiment;

[0096] Figure 9 Flowchart of the earthwork volume prediction model for the embodiment;

[0097] Figure 10 Optimizing the height prediction flow chart of the energy dissipation equipment for the embodiment;

[0098] Figure 11 A simple artificial neural network predictive model for optimizing the height of energy dissipation equipment;

[0099] Figure 12 Highly optimized system software module for the energy dissipation device on trailing suction hopper dredger;

[0100] Figure 13 Highly automated adjustment process of the energy dissipation device of the trailing suction hopper dredger during construction;

[0101] Reference numerals

[0102] 1- telescopic main pipe, 2- transverse pipe, 3- connecting steel plate, 4- hydraulic support rod, 5- mud tank inner beam, 6- mud tank side wall;

[0103] 11-inner pipe, 12-outer pipe, 13-filler, 14-limit block, 15-main pipe connecting flange;

[0104] 21-outlet, 22-transverse pipe connection flange;

[0105] 41-rear cylinder head, 42-cylinder barrel, 43-piston, 44-piston rod, 45-guide sleeve, 46-sealing ring, 47-front cylinder head. DETAILED DESCRIPTION

[0106] The adjustable height energy dissipation device, automatic height adjustment optimization method and system of the trailing suction hopper dredger of the present invention will be further described in detail below with reference to specific embodiments in conjunction with the accompanying drawings.

[0107] Preparation process

[0108] When the trailing suction hopper dredger is modified, the structural part of the adjustable height energy dissipation device is as follows: Figures 1 to 4 shown.

[0109] The energy dissipation device is height-adjustable and is embedded in and controlled by the shipboard loading control system, so as to make timely adjustments for optimal loading operations.

[0110] The adjustable height energy dissipation device of the trailing suction hopper dredger consists of a telescopic main pipe 1, a transverse pipe 2, a connecting steel plate 3, a hydraulic support rod 4, and a cross beam 5 in the mud chamber; the cross beam 5 in the mud chamber is fixed to the side wall 6 of the mud chamber by welding technology, and four lifting hydraulic support rods 4 are fixed to the cross beam 5 in the mud chamber by bolts. A connecting steel plate 3 is placed above the hydraulic support rod 4 and fixed to the upper end of the hydraulic support rod 4 by bolts. The connecting steel plate 3 is connected to the outer pipe 12 of the telescopic main pipe 1 by welding.

[0111] The telescopic main pipe 1 is a sleeve-like structure. Filler 13 seals the inner and outer pipes 11 and 12, preventing them from sliding relative to each other. Limit blocks 14 restrict the telescopic main pipe 1's extension and contraction. Hydraulic struts 4 allow the main pipe 1 to freely extend and retract within certain limits. Outer pipe 12 is connected to transverse pipe 2 via a main pipe connection flange 15.

[0112] The transverse pipe 2 is divided into three parts: left, middle and right, which are connected in sequence through flanges. The transverse pipe 2 is provided with a total of 8 outlets 21 for dissipating energy and loading the mud into the cabin.

[0113] The hydraulic strut 4 consists of a rear cylinder head 41, a cylinder barrel 42, a piston 43, a piston rod 44, a guide sleeve 45, a sealing ring 46, and a front cylinder head 47. The rear cylinder head 41 is bolted to the crossbeam 5 in the mud tank and seals the rear of the cylinder barrel 42. The piston 43 receives force within the cylinder barrel, which is transmitted to the connecting steel plate 3 above via the piston rod 44. The guide sleeve 45 supports and ensures the coaxiality of the piston rod 44 and the cylinder barrel 42. The sealing ring 46 seals the front cylinder head 46 and the piston rod 44 to prevent water, mud, and sand from affecting the normal operation of the components within the cylinder barrel 42 during loading.

[0114] When adjusting the height of the energy dissipation device, the ship-borne loading control system controls the lifting and lowering of the four hydraulic support rods 4 in the embodiment, and drives the outer pipe 12 and the transverse pipe 2 of the telescopic main pipe 1 to rise or fall through the connecting steel plate 3. The mud is discharged into the mud tank along the inner pipe 11, the outer pipe 12, the transverse pipe 2, and the outlet 21 to carry out the loading operation.

[0115] Example 1

[0116] Optimization method for automatic height adjustment of energy dissipation device of trailing suction hopper dredger

[0117] Process 1: Complete the facility modification of the trailing suction hopper dredger by referring to the preparatory work.

[0118] Process 2 determines the optimal adjustment height value, including the following steps:

[0119] Step 1: Determine the core parameters, including the key parameters of six parts: geometric model, boundary conditions, grid conditions, initial conditions, fluid motion and sediment motion, which are used to automatically establish the three-dimensional model of the mud tank system and set the scenario constraints of the loading simulation.

[0120] (1) Geometric model parameters, including the size of the mud tank structure, the size of the transverse and longitudinal triangular tanks, the size of the overflow equipment, the size of the energy dissipation equipment, and the position of the transverse and longitudinal triangular tanks and the energy dissipation overflow equipment relative to the mud tank, are used to establish a three-dimensional structural model of the mud tank system. Figure 5 shown.

[0121] (2) Boundary conditions include setting the inlet boundary as the loading flow rate and loading mud concentration, selecting the overflow boundary as the free outflow boundary, the upper boundary of the mud tank as the atmospheric pressure boundary, and the bulkhead boundary as the wall boundary with a certain roughness material;

[0122] (3) Figure 6 As shown, the grid is divided according to the three-dimensional model. The grid conditions include setting the total grid range in three directions to include the mud tank system structure model, and setting the unit grid size to the appropriate length, width and height dimensions for identifying the mud tank structure in the simulation calculation;

[0123] (4) Initial conditions include setting the initial water level and loading simulation duration;

[0124] (5) Fluid motion parameters include setting fluid temperature and fluid viscosity, and selecting the fluid motion equation as the RNG k-ε equation;

[0125] (6) Sediment movement parameters include setting sediment particle size, sediment density, sediment porosity, critical Shields number, underwater angle of repose, displacement coefficient and scour coefficient. The empirical formula of Mastbergen and Von den Berg is used to calculate the sediment initiation and scour process. The empirical formula of equat i is used to calculate the bed load transport process during the simulation.

[0126] Step 2: Establish a loading numerical model and loading simulation prediction;

[0127] Establishing a numerical model for loading: This refers to using CFD software to establish a numerical model that can simulate the loading construction process based on the input core parameters to achieve loading simulation function;

[0128] Loading simulation prediction: Using the above loading numerical model, we set different energy dissipation device height positions and conduct numerical simulations to simulate the entire loading process. This allows us to simulate and predict the changes in loading volume, earthwork volume, total overflow loss, and sediment height in the cabin over loading time. We then obtain multiple sets of numerical simulation results corresponding to different energy dissipation device height positions. The energy dissipation device height position needs to be set manually.

[0129] The numerical simulation results also include: t (3D grid point cloud of flow at time t); ρ t (3D grid point cloud with density at time t); v x (3D grid point cloud of velocity in x direction at time t); v y (Three-dimensional grid point cloud of the y-direction velocity component at time t); the results of the numerical calculation are provided to step 3.

[0130] The embodiment uses CFD software, structured multi-block grid technology and finite volume method to automatically establish a numerical model of the loading construction process based on the input core parameters, and realizes the simulation of the loading construction changing with time. The simulation results are as follows: Figure 7 As shown in the figure, different energy dissipation device heights are set. The loading control system sets different sediment movement parameters based on the geological survey report of the construction area and references existing construction experience data for different soil types to perform loading simulation calculations under multiple working conditions.

[0131] Step 3: Energy dissipation device highly optimizes output.

[0132] The optimization goal is to optimize the sediment flatness α and maximize the earthwork volume D, then the objective function of the optimization model is:

[0133]

[0134] max D=max ∑ΔQ i ·Δt·ρ (16)

[0135] Set constraints (xt):

[0136] X = {x1,x2,x3…,x i} (17)

[0137] Among them, Δt is the length of the time period for calculating the amount of deposited earthwork; ΔQ i is the outflow flow rate in the time period Δt; ρ is the outflow concentration in the time period Δt;

[0138] x i Belongs to [0,H], H represents the height of the mud tank, x (it) As a parameter for calculating flatness, it represents the height of sediment deposited at sampling position i at time t, It represents the average height of sediment deposited in the longitudinal centerline section of the mud chamber.

[0139] t <T (18)

[0140] Where T represents the maximum time set initially.

[0141] Set the heights of multiple energy dissipation devices and use the numerical calculation model in step 2 to calculate q t , ρ t 、v x and v y , and then calculate the flatness of the sediment and the amount of earthwork. According to the above optimization objectives (optimal flatness and maximum earthwork), the best group is selected, and the corresponding energy dissipation equipment height is the optimal energy dissipation equipment height.

[0142] Alternatively, the recommended value for optimizing the height of the energy dissipator can be determined based on the evaluation criteria (first gradient, second gradient) of four indicators: load capacity, earthwork volume, overflow loss, and sediment flatness. Typically, loading efficiency, represented by load capacity, earthwork volume, and overflow loss, is used as the first gradient criterion, while sediment flatness is used as the second gradient criterion. The greater the load capacity and earthwork volume, the smaller the overflow loss, and the smoother the mud tank, the better the impact of the energy dissipator height on the loading efficiency under these conditions.

[0143] Furthermore, in construction conditions where higher requirements are placed on the loading flatness, the evaluation criteria order can be adjusted to take the deposited sediment flatness as the first gradient.

[0144] In step 3, to avoid complex numerical calculations, a neural network is introduced to predict sediment height and earthwork volume.

[0145] Combine Figure 8 As shown in the figure, four variables are input into the input layer: qt (flow), ρt (density), vxt (x-direction velocity) and vyt (y-direction velocity). These input features are preprocessed through feature selection and standardization and then input into the target Transformer neural network model to obtain the sediment height prediction results.

[0146] The target Transformer neural network model includes an Embedding Layer, Multi-head Attention, and Softmax (probability distribution output). The Embedding Layer vectorizes the input features and converts them into vector representations suitable for network processing. This process maps high-dimensional input features to a low-dimensional vector space for subsequent processing. Multi-head Attention captures different feature information through multiple sets of attention heads, integrates this information, and performs weighted processing on the input features. Softmax calculates the probability distribution of each output based on the results processed by the multi-head attention mechanism:

[0147]

[0148] The output value is mapped to [0,1], and the sum of all output probabilities is 1, which helps the model learn the nonlinear relationship between input features and output results. Finally, the predicted sediment deposition height sequence (x1, x2, ..., x n ).

[0149] The Transformer neural network model (sedimentation and sediment height prediction model) is trained using four input variables as training data and sedimentation height sequences as labels. After each iteration, the loss function is calculated using validation data and its labels, with mean square error (MSE) as the loss function:

[0150]

[0151] Where x i ,x i are the sediment height obtained by iterative calculation and the actual sediment height as the label value, respectively.

[0152] The model adjusts the network weights through the backpropagation algorithm to minimize the loss function. In each training iteration, the gradient of the loss function is calculated and the network weights are updated until the model reaches the preset convergence condition, resulting in the target neural network model.

[0153] Combine Figure 9 As shown in the figure, four variables are input into the input layer: qt (flow), ρt (density), vxt (x-direction speed), and vyt (y-direction speed). The input features are preprocessed (normalized) and then input into the network. The BatchNorm layer calculates the mean and variance of all values ​​in the mini-batch (small batch), and uses the calculated mean and variance to normalize each batch of data, so that the input of each layer maintains a relatively stable distribution, which is used to accelerate training and improve the stability of the model. The PointNet layer is a deep learning framework that specializes in processing point cloud data. In this model, it is used to extract high-dimensional feature representations of input features. The fully connected layer further processes the features extracted from the PointNet layer and can map the high-dimensional features to the target output space, thereby realizing the prediction of earthwork volume.

[0154] Optionally, the present invention also provides another simpler and faster method for recommending parameters of energy dissipation equipment height optimization, combining Figure 10 The figure shows the optimal energy dissipation equipment height prediction process. The prediction process includes 10 parts: data input, data preprocessing, training set and test set division, neural network model construction, forward propagation, loss function calculation, backpropagation, parameter update, model evaluation, and optimal value recommendation.

[0155] (1) Data input: As the starting point of the entire prediction model, the data of the mud entering the mud tank at time t, such as qt (flow rate), ρt (density), vxt (x-direction velocity), vyt (y-direction velocity), and h (height of the energy dissipation device), are input in tabular form.

[0156] (2) Data preprocessing: In this stage, the data described in (1) will be standardized. The earthwork volume D and sediment flatness α are calculated, and a new parameter λ (0 < λ < 1) is set to calculate λD-(1-λ)α for each set of data.

[0157]

[0158] D=∑ΔQ i ·Δt·ρ (4)

[0159] Where: Δt is the length of the time period for calculating the amount of deposited earthwork; ΔQ i is the outflow flow rate in the Δt time period; ρ is the outflow concentration in the Δt time period

[0160] (3) Training and test set division: Data such as qt (flow rate), ρt (density), vxt (x-direction velocity), vyt (y-direction velocity), and h (height of the energy dissipator) are divided into training and test sets. The training set is used to train the model, and the test set is used to evaluate the model's performance. The division ratio is set to 8:2 or 7:3.

[0161] (4) Neural network model construction: combined with Figure 11 The following figure shows a simple artificial neural network (MLP) model. A neural network consists of multiple layers of neurons (nodes), with each layer containing several neurons. The layer structure includes an input layer (which receives processed input data), hidden layers (which perform complex calculations and extract high-level features of the input data), and an output layer (which produces the final prediction result, i.e., the objective function λD-(1-λ)α).

[0162] (5) Forward propagation: Data passes through the neurons in each layer from the input layer, calculates the activation function and outputs it until it reaches the output layer. The calculation formula for each layer is as follows:

[0163] z (l) =W (l) a (l-1) +b (l) (5)

[0164] a (l) =σ(z (l) ) (6)

[0165] Where: l is the number of activation layers; W (l) is the weight matrix; b (l) is the bias vector; a (l) with a (l-1) are the activation values ​​of the current layer and the previous layer respectively. In the input layer a, the corresponding input data qt (flow), ρt (density), vxt (x-direction speed) and vyt (y-direction speed), h (height of energy dissipation device) can be taken for calculation; σ is the activation function.

[0166] Furthermore, the activation function selects the ReLU function:

[0167] σ(z (l) )=ReLU(z (l) )=max(0,z (l) ) (7)

[0168] (6) Loss function calculation: The common mean square error (MSE) is selected as the loss function:

[0169]

[0170] Where: y i is the true value, y i is the predicted value, where y = λD-(1-λ)α and n is the number of samples.

[0171] (7) Backpropagation: By calculating the gradient of the loss function with respect to each parameter, the parameters are updated to minimize the loss function. The gradient calculation formula is as follows:

[0172]

[0173]

[0174] Where: δ (l) is the error term, which is the difference between the network prediction value and the true label value.

[0175] (8) Parameter update: Use the gradient descent optimization algorithm to update the weight matrix and bias vector. The gradient descent update formula is as follows:

[0176]

[0177]

[0178] Where: α is the learning rate.

[0179] (9) Model evaluation: Use the test set to evaluate model performance using indicators such as accuracy, precision, and recall.

[0180] (10) Optimal value recommendation: After the model training is completed, predictions can be made based on the input parameters qt (flow), ρt (density), vxt (x-direction velocity), vyt (y-direction velocity), h (energy dissipation device height), etc. In order to find the maximum value of the model, the model parameters are fixed, and the chain rule is used to calculate the derivative of each parameter in each layer to make it equal to zero. The extreme value candidates within a certain range are obtained, and the maximum value is selected. The optimal energy dissipation device height h is obtained by backpropagation.

[0181] Process 3: Loading at the optimal energy dissipation height

[0182] Step 4: According to the optimal energy dissipation device height obtained in step 3, adjust the hydraulic support rod of the adjustable height energy dissipation device to raise or lower the height of the energy dissipation device to the optimal height.

[0183] Example 2

[0184] A height optimization system for energy dissipation devices on board a trailing suction hopper dredger

[0185] It includes 4 developed software modules: core parameter input module, loading process simulation prediction module, design optimization output module, and height automatic adjustment module. These software modules are embedded in the ship-borne loading control system and are called by it; Figure 12 ;

[0186] Among them, the core parameter input module:

[0187] The core parameters include geometric model, boundary conditions, grid conditions, initial conditions, fluid motion and sediment motion; the geometric model includes: mud tank structure size, transverse and longitudinal triangular tank size, overflow equipment size, energy dissipation equipment size and height, and the position of transverse and longitudinal triangular tanks and energy dissipation overflow equipment relative to the mud tank; the boundary conditions include the inlet boundary as the loading flow and loading mud concentration, the overflow boundary as the free outflow boundary, the upper boundary of the mud tank as the atmospheric pressure boundary, and the bulkhead boundary as the wall boundary with a certain roughness material; the grid conditions include: grid range in the x, y, and z directions and unit grid size; the initial conditions include the initial water level H0 and the loading simulation time T; the fluid motion parameters include fluid temperature T, fluid viscosity μ, and fluid motion equation; the sediment motion parameters mainly include sediment particle size, sediment density, sediment porosity, critical Shields number, underwater repose angle, push motion coefficient and scour coefficient, involving four forms of motion: sediment suspension, sediment deposition, sediment scour and sediment push;

[0188] Among them, the loading process simulation prediction module:

[0189] The core parameter input module uses the various parameters input to perform numerical calculations and simulate the loading process under different working conditions. The loading process simulation prediction module implements loading process simulation prediction, including the establishment of a loading numerical model and loading simulation prediction.

[0190] Establishing a numerical model for loading: refers to establishing a numerical model capable of simulating the loading construction process based on the input core parameters using CFD software (existing technology) to achieve loading simulation function.

[0191] Loading simulation prediction: Using the aforementioned loading numerical model, we perform numerical simulations at different energy dissipation device heights. This simulates the entire loading process and predicts how the loading volume, earthwork volume, total overflow loss, and sediment height within the tank change over time. Multiple sets of numerical simulation results are obtained for different energy dissipation device heights. The energy dissipation device height must be set manually.

[0192] The numerical simulation results also include: t (3D grid point cloud of flow at time t); ρ t (3D grid point cloud with density at time t); v x (3D grid point cloud of velocity in x direction at time t); v y (3D grid point cloud of the y-direction velocity component at time t). The results of the numerical calculation are provided to the design optimization output module.

[0193] This module displays various calculation data in real time and visualizes simulation results within a 3D model. By viewing outputs such as specific flow rates, concentrations, and earthwork volumes during loading under various conditions, users gain a more intuitive and in-depth understanding of the loading process. This helps users analyze various loading phenomena, such as flow rate distribution and sediment concentration variations, and optimize loading plans.

[0194] Among them, the design optimization output module:

[0195] Based on the numerical simulation results output by the loading process simulation prediction module, the evaluation criteria are set in order using four indicators, including loading capacity, earthwork volume, overflow loss, and sediment flatness, to obtain the optimized recommended height of the energy dissipation device; users can click to select the order of evaluation criteria (loading efficiency, flatness) according to actual needs; the system will automatically output the optimized recommended value for the height of the energy dissipation equipment based on the evaluation criteria selected by the user.

[0196] Specifically, in multiple sets of numerical simulation results, different energy dissipation device heights correspond to different four indicators: loading capacity, earthwork volume, overflow loss, and sediment flatness. An evaluation basis sequence is set, and the recommended value for the optimal design of the energy dissipation device height is determined based on the evaluation basis sequence (first gradient, second gradient). Usually, the loading efficiency represented by loading capacity, earthwork volume, and overflow loss is used as the first gradient judgment standard, and the sediment flatness is used as the second gradient judgment standard. Among them, the larger the loading capacity and earthwork volume, the smaller the overflow loss, and the higher the mud tank flatness, indicating that the energy dissipation device height in this working condition has a better impact on the loading effect.

[0197] Among them, the height automatic adjustment module:

[0198] Based on the recommended value for the optimized height of the energy dissipation device output by the design optimization output module, a control signal is sent via the shipboard loading control system to automatically control the hydraulic struts on the adjustable height energy dissipation device. By precisely controlling the extension and retraction of the hydraulic struts, the height of the energy dissipation device can be adjusted automatically and precisely. In addition, users can also manually determine the soil conditions of the construction area through existing geological survey data and input the appropriate height of the energy dissipation device through the loading control system based on engineering experience. After receiving the command, the system will send a control signal to the hydraulic struts to achieve manual auxiliary lifting and lowering adjustment of the energy dissipation device. This adjustment method is more flexible and can adapt to different construction environments and needs.

[0199] The modules are as follows:

[0200] The loading process simulation and prediction module uses the parameters entered in the core parameter input module to numerically simulate the loading process. Treating the sediment as solid particles, a fluid model and sediment particle model are used to simulate four states: suspension, deposition, resuspension, and propulsion. Using a block-structured grid, the energy dissipation device outlet is individually partitioned and meshed twice as densely. The inlet boundary is the loading flow rate and the loading mud concentration; the overflow boundary is the free outflow boundary; the upper boundary of the mud tank is the atmospheric pressure boundary; and the bulkhead boundary is a wall with a certain roughness. The finite volume method is used to numerically simulate the loading process.

[0201] During the simulation process, the module displays various calculation data in real time, such as flow rate, concentration, and earthwork volume, and visualizes the simulation results within the 3D model. This intuitive data and images provide users with a deeper and more comprehensive understanding of the loading process. Furthermore, by reviewing the output results under various operating conditions, users can analyze various loading process phenomena, such as flow rate distribution and sediment concentration changes, to further optimize the loading plan.

[0202] Furthermore, the loading capacity, earthwork volume, overflow loss, and sediment flatness at different times during the loading process are calculated using the following formula:

[0203] Loading capacity: refers to the total mass of mud mixture loaded into the mud tank without overflowing.

[0204] M=M f +M ss +M sd

[0205] Where: M represents the load, M f Indicates the mass of the fluid in the cabin, M ss Indicates the mass of suspended sediment particles in the cabin, M sd Indicates the mass of sediment particles deposited in the cabin.

[0206] Earthwork volume: refers to the volume of original soil loaded into the mud tank without overflowing.

[0207]

[0208] Where, ρ s represents the density of sediment particles, e represents the porosity of the original soil, ρ f represents the fluid density, ρ s1 represents the original soil density.

[0209] Overflow loss: refers to the total mass of mud mixture loaded into the mud tank but overflowed.

[0210] M l =M o -M

[0211] Where M o It is the total mass of the mud mixture flowing out from the outlet of the energy dissipation equipment.

[0212] Flatness inside the cabin: expressed by the standard deviation of the height of sediment deposited in the section along the longitudinal centerline of the mud cabin.

[0213]

[0214] Where α represents the flatness of the cabin, x i represents the height of sediment at position i on the longitudinal centerline of the mud tank, It represents the average height of sediment deposited on the longitudinal centerline section of the mud tank, and n represents the number of sediment deposited heights calculated on the longitudinal centerline section of the mud tank.

[0215] The automatic height adjustment module is a concrete manifestation of the shipboard loading control system's automation, eliminating the need for manual adjustment of the energy dissipation device's height. Based on the recommended optimized height of the energy dissipation device obtained by the design optimization output module, this module automatically controls the hydraulic struts (4) installed on the adjustable height energy dissipation device. By precisely controlling the extension and retraction of the hydraulic struts (4), the height of the energy dissipation device can be adjusted in real time, automatically, and with high precision. This automatic adjustment method not only improves accuracy and efficiency, but also reduces the difficulty and error of manual operation.

[0216] Furthermore, users can also input the appropriate energy dissipation device height through the shipboard loading control system based on their own judgment and experience. Upon receiving the command, the system sends a corresponding control signal to the hydraulic strut 4, enabling manual adjustment of the energy dissipation device. This flexible adjustment method can adapt to different construction environments and requirements, improving the system's adaptability and practicality.

[0217] Example 3

[0218] A new type of trailing suction hopper dredger with automatically adjustable energy dissipation device height and its loading construction method: by modifying the original trailing suction hopper dredger or building a new ship, the hardware is equipped with an energy dissipation device with adjustable height, and the preparatory work is done, which will not be repeated in the previous article.

[0219] The height-adjustable energy dissipation device is managed by the shipboard loading control system, which has been introduced in Example 2.

[0220] Loading operation method for highly automated adjustment of energy dissipation device of trailing suction hopper dredger

[0221] Specific process, such as Figure 13 , including the following steps:

[0222] Step 1: The trailing suction hopper dredger enters the construction area; before construction begins, core parameters are manually input, mainly including six parts: geometric model, boundary conditions, grid conditions, initial conditions, fluid motion and sediment motion. The specific parameters are as described in Example 1.

[0223] Step 2: The shipboard loading control system automatically establishes a numerical model of the loading construction process based on the input core parameters, realizing the simulation function of the loading construction changing over time.

[0224] Step 3: Set different energy dissipation equipment heights in the shipboard loading control system and perform loading simulation calculations under multiple working conditions.

[0225] Step 4: Recommend energy dissipation equipment height parameters through the design optimization parameter output module of the shipboard loading control system. Select the evaluation criteria sequence (first gradient, second gradient) and the number of recommended values ​​according to user requirements to automatically generate optimized design values ​​for the energy dissipation equipment height. The core parameters for energy dissipation equipment height position recommendations are based on the default first gradient criteria for loading efficiency, represented by load capacity, earthwork volume, and overflow loss, and the second gradient criteria for sediment flatness.

[0226] Furthermore, in construction conditions where higher requirements are placed on the loading flatness, the evaluation criteria order can be adjusted to take the deposited sediment flatness as the first gradient.

[0227] Step 5: When the location and excavation depth of the ship construction area change significantly, refer to the geological survey report and the corresponding excavation soil changes. Based on the updated input values ​​of qt, ρt, vxt and vyt, repeat step 4 to recommend new optimized values ​​for the height of the energy dissipation equipment.

[0228] Specifically, as excavation progresses, the excavation position and depth change, and the soil conditions change. To improve the overall loading efficiency, the loading control system changes the sediment movement parameters based on the geological survey data, re-performs the numerical calculation and neural network training process such as loading simulation prediction under multiple working conditions, automatic optimization and recommendation of height parameters, and the loading control system automatically re-controls the height of the hydraulic support rod, thereby adjusting the height of the energy dissipation device to always ensure high loading efficiency. When the amount of fine particles in the excavated soil increases, the height of the energy dissipation device is automatically and accurately lowered to reduce the disturbance of the inflowing water to the fine particles of sediment, allowing the fine particles to settle quickly and reducing the overflow loss during loading. When the amount of coarse particles in the loading mud increases, the particles settle quickly, and the height of the energy dissipation device is automatically and accurately increased to prevent soil particles from blocking the outlet 21 of the energy dissipation device.

[0229] Step 6: When the trailing suction hopper dredger is operating at different construction locations and excavation depths, the above-mentioned corresponding energy dissipation equipment height optimization recommended value is called in combination with the ship position and the lowering depth of the rake arm, and a command is sent to the hydraulic support rod 4 set on the adjustable height energy dissipation device to realize real-time height adjustment of the energy dissipation device.

Claims

1. A method for optimizing the height automatic adjustment of the energy dissipation device of a trailing suction hopper dredger, characterized in that: The following steps are involved: Process 1: Constructing an adjustable height energy dissipation device in the cabin of the suction dredger; Process 2 determines the optimal energy dissipation height; it includes the following steps: Step 1: Determine core parameters using existing trailing suction hopper construction data, geological survey reports of the construction area, and soil data, including: geometric model, boundary conditions, grid conditions, initial conditions, fluid motion, and sediment movement; Step 2: Establish a loading numerical model and loading simulation prediction; Establishing a numerical model for loading: This refers to using CFD software to establish a numerical model that can simulate the loading construction process based on the input core parameters to achieve loading simulation function; Loading simulation prediction: Using the above loading numerical model, we set different energy dissipation device height positions and conduct numerical simulations to simulate the entire loading process. This allows us to simulate and predict the changes in loading volume, earthwork volume, total overflow loss, and sediment height in the cabin over loading time. We obtain multiple sets of numerical simulation results corresponding to different energy dissipation device height positions. The energy dissipation device height position needs to be set manually. The numerical simulation results also include: the three-dimensional grid point cloud of the flow at time t, denoted by q t ; The three-dimensional grid point cloud with density at time t, record ρ t ; t time The three-dimensional grid point cloud of the direction velocity, v x ; The three-dimensional grid point cloud of the y-direction velocity at time t, recorded v y ;The results of numerical calculation are provided to step 3; Step 3: Based on the numerical simulation results, the evaluation criteria are set according to the four indicators of loading capacity, earthwork volume, overflow loss, and sediment flatness to obtain the optimized recommended height of the energy dissipation device; Process 3: Loading at the optimal energy dissipation height Step 4: According to the optimal energy dissipation equipment height value obtained in step 3, raise or lower the height of the adjustable height energy dissipation device to the optimal height.

2. The method according to claim 1, wherein: Step 1: The RNG k−ε equation is used to describe the fluid motion according to the loading characteristics; the empirical formula of Mastbergen and Von den Berg is used to calculate the sediment initiation and scouring process.

3. The method according to claim 1, wherein: In step 3, in multiple sets of numerical simulation results, different energy dissipation device heights correspond to different loading capacities, earthwork volumes, overflow losses, and sediment flatness indicators. The recommended value for the optimal design of the energy dissipation device height is determined according to the evaluation criteria sequence. The evaluation criteria sequence includes a first gradient and a second gradient. Usually, the loading efficiency represented by the loading capacity, earthwork volume, and overflow loss is used as the first gradient judgment standard, and the sediment flatness is used as the second gradient judgment standard. Among them, the greater the loading capacity and earthwork volume, the smaller the overflow loss, and the higher the mud tank flatness, the better the impact of the energy dissipation device height on the loading effect under this working condition.

4. The method according to claim 3, wherein: The optimization goal is to optimize the sediment flatness α and maximize the earthwork volume D, then the objective function of the optimization model is: Set constraints ( . t .): in, The length of the time period for calculating the amount of deposited earthwork; For Outflow volume within the time period; For Outflow concentration during the time period; i Belongs to [0,H], H represents the height of the mud tank, it As a parameter for calculating flatness, it represents the height of sediment deposited at sampling position i at time t, It represents the average height of sediment deposited on the longitudinal centerline section of the mud chamber; Where, T represents the maximum time of initial setting; Set the heights of multiple energy dissipation devices and use the numerical calculation model in step 2 to calculate q t 、 ρ t 、 v x and v y , and then calculate the flatness of the sediment and the amount of earthwork, optimize the goals based on the optimal flatness and maximum earthwork volume to select the best group, and the corresponding energy dissipation equipment height is the optimal energy dissipation equipment height.

5. The method according to claim 4, wherein: The number of energy dissipation device height groups is 12 to 20, with an interval of 0.25 to 0.5 m between each group, and a height variation range of 5 to 6 m.

6. The method according to claim 4, wherein: The neural network is introduced to predict sediment height and earthwork volume as follows: A sediment height prediction model is established to predict sediment height as follows: The calculation is done by establishing The predicted model of (t) value and initial and boundary conditions of numerical simulation is as follows: Input value: q t is the three-dimensional grid point cloud of the flow at time t; ρ t is the three-dimensional grid point cloud with density at time t; v x is time t 3D grid point cloud of directional velocity components; v y is the three-dimensional grid point cloud of the y-direction velocity at time t; Output value: (t+Δt) represents the sediment height matrix of the flatness section calculated at the next time Δt, which is used as the label item of the sediment height prediction model. The three-dimensional positions of the energy dissipation equipment are all set to 0. This setting characterizes the spatial characteristics of the energy dissipation equipment. The q values ​​calculated by numerical simulation are extracted at time intervals of Δt. t , ρ t 、 v x 、 v y q corresponding to each time t , ρ t 、 v x 、 v y Convolution kernel features are extracted from the three-dimensional grid data to generate a feature matrix. The corresponding values ​​used to calculate the sediment height in the cabin at time t+Δt are combined into a label matrix X to train the network. The sediment height prediction model for the calculation points of the flatness calculation plane obtained through the above operation is trained using numerical simulation data. During the training phase, the input and label are calculated by the loading process simulation prediction model in step 2; after the training is completed, the actual q t , ρ t 、 v x 、 v y Predict sediment height; The earthwork volume prediction model is established to predict the earthwork volume in the cabin as follows: As above, dt is the three-dimensional grid point cloud of earthwork volume at time t, and d(t+Δt) represents the earthwork volume calculated at the next Δt time as the label item of the earthwork volume prediction model; establish q at time t t , ρ t 、 v x 、 v y The earthwork volume prediction model is obtained by training a network based on the data and the three-dimensional data of the earthwork volume in the mud tank at time t+Δt.

7. The method according to claim 1, wherein: An adjustable height energy dissipation device based on a trailing suction hopper dredger can be modified on the original ship or built on a new ship; The height-adjustable energy dissipation device is designed to include a retractable main pipe (1), a transverse pipe (2), a connecting steel plate (3), a hydraulic support rod (4), and a crossbeam (5) in the mud chamber; The retractable main pipe (1) is a sleeve-type structure. The retractable main pipe (1) comprises an inner pipe (11), an outer pipe (12), a filler (13), a stopper (14), and a main pipe connecting flange (15). The inner pipe (11) and the outer pipe (12) are sealed by the filler (13). The stopper (14) limits the retractable range of the retractable main pipe (1). The inner pipe (11) is connected to the mud delivery pipeline of the trailing suction dredger, and the outer pipe (12) is connected to the transverse pipe (2) via the main pipe connecting flange (15). The transverse pipe (2) is divided into three parts: left, middle and right, which are connected in sequence through the transverse pipe connecting flange (22) and are provided with a total of 8 outlets (21); The mud tank inner crossbeam (5) is fixed to the mud tank side wall (6) by welding technology, and four hydraulic support rods (4) for lifting are fixed to the mud tank inner crossbeam (5) by bolts. A connecting steel plate (3) is placed above the hydraulic support rods (4) and fixed to the upper end of the hydraulic support rods (4) by bolts. The connecting steel plate (3) is connected to the outer pipe (12) of the telescopic main pipe (1) by welding technology; Four hydraulic support rods (4) are evenly arranged on the cross beam (5) in the mud tank, so as to provide sufficient force while ensuring that the cross beam (5) and the connecting steel plate (3) in the mud tank are evenly stressed; The hydraulic support rod (4) is composed of a rear cylinder cover (41), a cylinder (42), a piston (43), a piston rod (44), a guide sleeve (45), a sealing ring (46), and a front cylinder cover (47); the rear cylinder cover (41) is connected to the cross beam (5) in the mud tank by bolts, and at the same time closes the rear part of the cylinder (42); the piston (43) is subjected to force in the cylinder, and the force is transmitted to the upper connecting steel plate (3) through the piston rod (44); the guide sleeve (45) supports and ensures the coaxiality of the piston rod (44) and the cylinder (42); the front cylinder cover (47) and the piston rod (44) are sealed by the sealing ring (46) to prevent water and mud from affecting the normal operation of the components in the cylinder (42) during the loading process.

8. The method according to claim 7, wherein: Obtain the optimized recommended height of the energy dissipation device, adjust the hydraulic support rod of the adjustable height energy dissipation device, raise or lower the height of the energy dissipation equipment to the optimal height, and perform loading operations at the optimal energy dissipation height.

Images