Method for predicting the diffusion of a plume in seabed mining based on numerical simulation
By employing 3D terrain construction and multi-source data assimilation techniques, combined with dynamic source term modeling and multiphase flow coupling simulation, the accuracy problem of predicting plume diffusion in seabed mining was solved. This enabled high-precision 3D dynamic diffusion simulation and ecological risk zone delineation, thereby improving the environmental safety of mining operations.
Patent Information
- Application Number
- CN202610581997.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-04-29
- Publication Date
- 2026-07-24
AI Technical Summary
Existing technologies struggle to accurately predict plume diffusion caused by seabed mining activities in complex seabed topography and stratified water environments. Furthermore, they lack systematic integration of dynamic source term injection, particle size distribution modeling, and identification of ecologically sensitive areas, resulting in simulation results that deviate from reality and have poor real-time performance.
By employing 3D terrain construction, multi-source data assimilation, dynamic source term modeling, multiphase flow coupling simulation, and intelligent risk area identification, and combining historical and on-site multi-source ocean observation data with the 4D-Var algorithm, a high-precision initial flow field is constructed. Through multi-scale CFD simulation and Eulerian-Lagrange particle tracking, the 3D dynamic diffusion process of the plume is simulated.
It enables accurate modeling of plume diffusion processes in complex seabed topography and stratified water environments, improving simulation and prediction accuracy and adaptability to real working conditions, and providing environmental safety assurance for deep-sea mining operations.
Smart Images
Figure CN122452332A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of numerical simulation technology, and in particular to a method for predicting the diffusion of plumes in seabed mining based on numerical simulation. Background Technology
[0002] With the rapid advancement of deep-sea mineral resource development, the potential impacts of seabed mining activities on marine ecosystems have attracted widespread attention. In particular, during mining operations, particle plumes generated by equipment emissions can spread through the water, leading to localized water quality deterioration, sediment cover, and habitat disturbance. Therefore, accurate prediction of plume diffusion behavior is a core element of seabed mining environmental impact assessment and risk control. Traditional prediction methods are mostly based on two-dimensional empirical models or static profile observations, which are difficult to adapt to the coupled changes of complex topography, water stratification, and dynamic equipment movement. These methods suffer from low resolution, poor real-time performance, and large prediction errors. Furthermore, existing numerical simulation platforms generally lack systematic integration of dynamic source term injection, particle size distribution modeling, particle-fluid bidirectional coupling, and dynamic identification of ecologically sensitive areas, making it difficult to support a precise trade-off between mining scheduling and ecological protection.
[0003] Existing plume diffusion prediction methods generally suffer from several key technical shortcomings. On the one hand, in complex terrain and stratified water environments, traditional simplified models cannot effectively characterize the actual flow field boundary conditions, often neglecting the influence of seabed roughness, surface wind stress, and open boundary disturbances on the flow field evolution, leading to simulation results that deviate from actual observations. On the other hand, the processing of particulate source terms lacks physical constraints, often being set as fixed values or static distributions, failing to dynamically adjust the injection rate and direction according to the operating parameters of mining equipment, making it difficult to reflect actual emission behavior. Existing simulation systems still rely on static images or expert experience judgments for risk zoning output, lacking an automated dynamic zoning mechanism based on the spatiotemporal distribution of concentration, making it difficult to support real-time decision-making and the protection needs of ecologically sensitive areas during the mining process. Summary of the Invention
[0004] This invention provides a method for predicting the diffusion of plumes in seabed mining based on numerical simulation. It integrates key technologies such as multi-source data assimilation, complex boundary processing, dynamic source term modeling, multiphase flow coupling simulation, and intelligent identification of risk areas. Based on three-dimensional terrain construction and abrupt layer identification, it uses the 4D-Var algorithm to fuse historical and on-site multi-source marine observation data to construct a high-precision initial flow field.
[0005] A numerical simulation-based method for predicting the diffusion of plumes in seabed mining includes the following steps: S1, 3D Topographic Field Construction: Acquire seabed topographic measurement data of the target sea area, construct a digital topographic model, and generate the corresponding 3D mesh; S2, Environmental flow field boundary setting: Based on the constructed terrain model, historical ocean observation data and on-site sensor data are integrated to generate initial and boundary conditions for the environmental flow field, including flow velocity, flow direction and temperature-salinity structure, as the input flow field for plume simulation; S3, Particle Source Parameter Injection: Combining the emission characteristics of seabed mining equipment, a plume flow source term is set, including particle size distribution, mass concentration, emission rate and direction information, and the plume flow source term is injected into the flow field boundary to drive particle diffusion simulation; S4, Three-dimensional diffusion numerical simulation of plume flow: Based on a multi-scale CFD solver, the terrain mesh of S1, the flow field boundary of S2 and the particle source parameters of S3 are input to simulate the diffusion process of plume flow in three-dimensional space and output the spatial distribution map of particle concentration field evolution over time. S5, Dynamic Impact Domain Identification and Risk Zone Delineation: Based on the output spatial distribution map and combined with the setting of ecological sensitivity thresholds, the plume diffusion impact area is automatically delineated, high-risk, medium-risk and low-risk areas are identified, and corresponding dynamic risk maps are generated.
[0006] Optionally, S1 includes: S11, Acquisition of seabed topographic measurement data: Collect seabed topographic point cloud data of the target sea area through multibeam echo sounder, side-scan sonar or lidar equipment to obtain a set of discrete coordinate points; S12, Terrain data preprocessing: Based on the acquired discrete coordinate point set, statistical screening of water depth data is performed to remove outliers. Spatial interpolation is then performed on the removed point set to construct a water depth function. S13, Construction of digital terrain model: Based on the Kriging interpolation function, a digital elevation model in the form of a regular network is constructed; S14, 3D Network Generation: The digital elevation model is divided into three dimensions, and the Delaunay triangulation algorithm is used to generate an unstructured tetrahedral mesh set.
[0007] Optionally, S2 includes: S21, Multi-source ocean data fusion: acquire historical ocean observation datasets and real-time datasets from on-site sensors, fuse the above data based on a four-dimensional variational assimilation algorithm, and construct the optimal initial state field; S22, Initialization of 3D environmental flow field: Map the optimal state field to a 3D unstructured mesh to construct the initial physical field; S23, Dynamic boundary condition setting: Based on the hydrodynamic characteristics of the seabed mining area, appropriate boundary conditions are applied to different boundary types of the numerical simulation domain. S24, Temperature-Salinity Boundary Constraint Setting: A relaxation forcing term is introduced at the open boundary to control the boundary temperature and salinity to approach the observed values.
[0008] Optionally, S22 includes: S221 maps the optimal state field to a three-dimensional unstructured network to construct a three-dimensional velocity field; S222 maps the optimal state field to a three-dimensional unstructured network to construct a three-dimensional temperature field and a three-dimensional salinity field.
[0009] Optionally, S23 includes: S231, Lateral Open Boundary Treatment: Radiation boundary conditions are used to treat the lateral boundaries of the numerical simulation region; S232, Sea Surface Boundary Modeling: Apply wind stress boundary conditions at the top edge of the simulation domain to simulate the influence of surface wind on the underlying fluid. The magnitude of wind stress is determined by air density, wind speed, and drag coefficient, while also considering turbulent viscosity effects. S233, Seabed Boundary Modeling: At the bottom edge of the simulation domain, the velocity distribution is approximated by a logarithmic wall function. Seabed roughness and friction velocity factors are introduced to reproduce the velocity profile characteristics of the plume near the seabed.
[0010] Optionally, S3 includes: S31, Emission Characteristic Parameterization: Based on the operating parameters of the target mining equipment, determine the emission attributes of the plume source term, including emission rate and emission direction vector; S32, Particulate matter size distribution modeling: The Rosin-Rammler distribution function is used to model the size distribution of multi-component particles and divide the particle size intervals; S33, Source Term Spatial Injection: Define the time-varying emission point locations within a three-dimensional mesh domain to construct a feather source term function; S34, Multiphase Flow Coupling Drive: Input the source term function into the Euler-Lagrange multiphase flow model.
[0011] Optionally, S31 includes: S311, Emission Rate Determination: Calculate the mass emission rate of the plume source based on the structural parameters and operating status of the target mining equipment; S312, Emission Direction Vector Calculation: Combining the actual attitude information of the target mining equipment, including pitch and yaw angles, the three-dimensional direction of particulate matter emission is determined, and the attitude angles are converted into direction vectors using trigonometric function relationships.
[0012] Optionally, S4 includes: S41, Solving the multi-scale CFD governing equations: Based on the three-dimensional terrain mesh generated in S1 and the environmental flow field boundary conditions set in S2, solve the unsteady Reynolds-averaged Navier-Stokes equations. S42, Particle Phase Diffusion Tracking: Based on the source term injected by S3, the evolution of the center and velocity of a single particle is tracked within the Eulerian-Lagrange framework; S43, Dynamic Reconstruction of Concentration Field: The mass of all particles per unit volume is statistically analyzed on the grid to construct a particle concentration field; S44, Spatiotemporal evolution data output: Outputs a three-dimensional concentration field sequence, with output formats including isosurface plots and profile cloud maps.
[0013] Optionally, S5 includes: S51, Threshold setting: Preset ecological sensitivity thresholds, including high-risk threshold, medium-risk threshold and low-risk threshold; S52, based on the output spatial distribution map and combined with the ecological sensitivity threshold setting, automatically divides the area affected by the plume diffusion and identifies high-risk, medium-risk and low-risk areas. S53. Based on the identified areas affected by the plume diffusion, a corresponding dynamic risk map is generated.
[0014] Optionally, S53 includes: S531, set a visual color scheme according to risk level; S532 generates dynamic risk maps, with output formats including: isosurface maps, profile cloud maps, and 3D heat maps.
[0015] The beneficial effects of this invention are: This invention provides a numerical simulation-based method for predicting the diffusion of plumes in seabed mining. It can accurately model the three-dimensional dynamic diffusion process of plumes in complex seabed topography and stratified water environments. By introducing a variational assimilation algorithm to fuse historical and real-time ocean data, it constructs a three-dimensional flow field boundary condition with strong physical consistency. Combining multi-scale CFD simulation with an Eulerian-Lagrange particle tracking framework, it accurately characterizes the spatiotemporal evolution of particles with multiple particle sizes. Finally, it generates a high-resolution dynamic concentration field sequence and can automatically delineate operational risk areas based on concentration thresholds.
[0016] This invention proposes innovative mechanisms in key aspects: First, it adopts a source term injection strategy with physical constraints to tightly couple the operating parameters of mining equipment with particle emission behavior, supporting dynamic direction control and particle size diversity modeling; Second, it introduces the particle cloud tracking method and the Gaussian kernel density estimation method to achieve continuous reconstruction of the concentration field under high computational efficiency; This improves the accuracy of simulation prediction and the adaptability to real working conditions, and also provides an important environmental safety guarantee foundation for deep-sea mining operations. Attached Figure Description
[0017] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only for this invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0018] Figure 1 This is a flowchart of the inventive method according to an embodiment of the present invention; Figure 2 This is a numerical simulation diagram of an embodiment of the present invention. Detailed Implementation
[0019] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments. It should also be noted that, to make the embodiments more comprehensive, the following embodiments are the best and preferred embodiments, and those skilled in the art can use other alternative methods to implement some well-known technologies; moreover, the accompanying drawings are only for more specific description of the embodiments and are not intended to specifically limit the present invention.
[0020] like Figures 1-2 As shown, the method for predicting the diffusion of seabed mining plumes based on numerical simulation includes the following steps: S1, 3D Topographic Field Construction: Acquire seabed topographic measurement data of the target sea area, construct a digital topographic model, and generate the corresponding 3D mesh; S2, Environmental flow field boundary setting: Based on the constructed terrain model, historical ocean observation data and on-site sensor data are integrated to generate initial and boundary conditions for the environmental flow field, including flow velocity, flow direction and temperature-salinity structure, as the input flow field for plume simulation; S3, Particle Source Parameter Injection: Combining the emission characteristics of seabed mining equipment, a plume source term is set, including particle size distribution, mass concentration, emission rate and direction information, and the plume source term is injected into the flow field boundary to drive particle diffusion simulation; S4, Three-dimensional diffusion numerical simulation of plume flow: Based on a multi-scale CFD solver, the terrain mesh of S1, the flow field boundary of S2 and the particle source parameters of S3 are input to simulate the diffusion process of plume flow in three-dimensional space and output the spatial distribution map of particle concentration field evolution over time. S5, Dynamic Impact Domain Identification and Risk Zone Delineation: Based on the output spatial distribution map and combined with the setting of ecological sensitivity thresholds, the plume diffusion impact area is automatically delineated, high-risk, medium-risk and low-risk areas are identified, and corresponding dynamic risk maps are generated.
[0021] S1 specifically includes: S11, Acquisition of Seabed Topographic Survey Data: High-resolution topographic scanning of the target sea area is performed using a multibeam echo sounder, side-scan sonar, or lidar equipment to acquire a discrete set of seabed topographic coordinates, represented as: ; in, For the discrete coordinate point set of the seabed topography, Indicates the first The coordinates of each measuring point on the horizontal plane (unit: meters). This indicates the water depth at the measuring point (unit: meters). This indicates the total number of data points collected. S12, Topographic Data Preprocessing: Processing Discrete Coordinate Point Sets of Seabed Topography The processing includes the following steps: (1) Outlier removal: The Raida criterion (3) is adopted. (Criteria) Remove data points that deviate significantly. Points that meet the following conditions will be removed, denoted as: ; in, It is the average water depth at all measuring points. It is the standard deviation; (2) Spatial interpolation processing: The spatial continuity of the removed data is reconstructed using the Kriging interpolation method, and a continuous water depth surface function is constructed, which is expressed as: ; ; in, For the first to participate in interpolation Water depth values at each node, The Kriging interpolation coefficients are controlled by a semivariogram and satisfy the nonnegativity constraint. and normalization constraints interpolation coefficients Linear equations controlled by semivariograms Please solve. Represents the spatial spacing vector. Indicates all spacing is A set of point pairs The total number of point pairs in the set. Points are used for interpolation. For index symbols in Kriging interpolation; S13, Terrain Digital Model Construction: Based on Interpolation Results Constructing a digital elevation model (DEM) on a regular grid is represented as follows: ; in, For the first Two-dimensional coordinates of a regular grid node, The interpolated water depth for the corresponding node. This represents the total number of nodes in the regular grid. S14, 3D Mesh Generation: The DEM is discretized in 3D using the Delaunay triangulation algorithm to generate an unstructured tetrahedral mesh set. It meets the following conditions: (1) The mesh element is a tetrahedral element. ,in, ; (2) All tetrahedrons satisfy the Delaunay criterion: their circumsphere does not contain any other mesh nodes, as shown below: ; in, The coordinates of the vertices of the tetrahedron. The center of the tetrahedron is the circumscribed sphere. Let the radius of the sphere be . The set of all grid nodes. Let be the coordinate vector of any node. This represents the total number of tetrahedral elements generated after the 3D mesh is generated.
[0022] S2 specifically includes: S21, Multi-source ocean data fusion: Acquiring historical ocean observation datasets and real-time datasets from on-site sensors, and fusing the data using a four-dimensional variational assimilation algorithm (4D-Var), specifically including: (1) Obtain historical ocean observation dataset, represented as: ; in, It is a historical ocean observation dataset. It is the first Coordinates of historical observation points (unit: m). The observation time is in seconds. It is the horizontal velocity vector (unit: ), Temperature (unit: ), It is the first Salinity (unit: PSU) at historical observation points. , It refers to the number of measuring points in historical ocean observation data; (2) Obtain the real-time dataset from the field sensors, represented as: ; in, It is a real-time dataset from on-site sensors. It is the first Coordinates of observation points (unit: m). The observation time is in seconds. It is the horizontal velocity vector (unit: ), Temperature (unit: ), It is the first Salinity at each observation point (unit: PSU) , The number of ocean data points collected by real-time sensors; (3) The above data are fused using the four-dimensional variational assimilation algorithm (4D-Var), and the optimization objective function is expressed as: ; in, It is a three-dimensional state variable vector, including the velocity field. Temperature field Salinity field , It is a dataset of historical ocean observations The generated background field, It is the background error covariance matrix. It is the first Similar to observational data (historical or real-time). It is the first Observational operators, It is the first Observational error covariance matrix; S22, Initialization of 3D Environmental Flow Field: Applying Optimal Assimilation Results Mapped to the constructed 3D mesh The initial field is generated, including a three-dimensional velocity field, a three-dimensional temperature field, and a salinity field, specifically including: (1) Three-dimensional velocity field: Unit: m / s; in, It is the three-dimensional initial velocity field vector. It is the east-west component of velocity (along the x-direction). It is the north-south velocity component (along the y-direction). It is the vertical component of velocity (along the z-direction); (2) Three-dimensional temperature field and three-dimensional salinity field: ,unit ; in, It is a three-dimensional temperature field. It is a three-dimensional salinity field. The coordinates of the three-dimensional mesh nodes; S23, Dynamic Boundary Condition Setting: Based on the hydrodynamic characteristics of the seabed mining area, appropriate boundary conditions are applied to different boundary types in the numerical simulation domain to ensure the physical rationality and stability of the plume diffusion simulation. Specifically, this includes: (1) Lateral open boundary: The lateral disturbance propagation is handled using the radiation boundary condition, expressed as: , ; in, Phase velocity (unit: ), The unit vector is the boundary normal direction. (2) Sea surface boundary: Wind stress is applied to the sea surface boundary, which is represented as: ; in, Turbulent viscosity coefficient (unit: The range of values is , It is the air density, with a typical value of approximately , It is the wind stress drag coefficient, with a value range of [value range missing]. , Sea surface wind speed (unit: ), It is sea level; (3) Seafloor boundary: The bottom flow velocity is approximated using a logarithmic wall function, expressed as: ; in, Friction speed (unit: ), Let be the von Kármán constant, taken as 0.4. It is the seabed roughness length (unit: m), and its value range is: ; S24, Temperature-Salinity Boundary Constraint Setting: A nudging term is introduced at the open boundary to constrain temperature and salinity changes, expressed as: ; in, These are boundary observations. It is the relaxation time constant (unit: s), which controls the regression rate of the field observations, and its value range is... .
[0023] S3 specifically includes: S31, Emission Characteristic Parameterization: Based on the operating parameters of the target mining equipment (collector), determine the emission attributes of the plume source term, including emission rate and emission direction vector, specifically including: (1) Emission rate : ; in, This refers to sediment density, in units of: , This refers to the cross-sectional area of the discharge outlet, in units of: , It is the discharge velocity, in units ; (2) Emission direction vector : via device pitch angle With yaw angle The calculation is expressed as: ; in, It is the pitch angle (unit: radians). Yaw angle (unit: radians); S32, Particulate Matter Size Distribution Modeling: The particle size distribution of multi-component particles is modeled using the Rosin-Rammler distribution function, dividing the particle size range into... Each particle size group is represented as: ; ; in, It is modeled using the Rosin-Rammler distribution function. Particle size (unit: m). Characteristic particle size (unit: m). It is the distribution index (dimensionless). It is the first Mass fraction of the group of particles For the first Upper and lower limits of particle size (unit: m); S33, Source Term Spatial Injection: Define the time-varying emission point locations within the three-dimensional mesh domain, construct the plume source term function, expressed as: ; ; in, It is the location of the source point. Let f be the source term function of the feathered flow, representing the first term. Particle size group in spatial position and time Injection intensity, Dirac function (unit: ), Equivalent radius of the discharge outlet (unit: m); S34, Multiphase Flow Coupling Drive: This will convert the source term... When input into the Eulerian-Lagrange multiphase flow model, the governing equations for the granular phase are expressed as: ; in, Particle velocity (unit: ), Fluid velocity (unit: ), This is the particle relaxation time (unit: s), with a range of values. ~10, Fluid dynamic viscosity coefficient (unit: Pa) ), range of values , It is the vector of gravitational acceleration (unit: ), Fluid density (unit: ).
[0024] S4 specifically includes: S41, Solving the multi-scale CFD governing equations: Based on the 3D terrain mesh generated by S1 Given the environmental flow field boundary conditions set by S2, the unsteady Reynolds-averaged Navier-Stokes equations (URANS) are solved, as follows: ; in, It is the fluid density. The temperature and salinity of S2 are calculated using the equation of state. It is the fluid velocity vector , It is fluid pressure , It is the total stress tensor. It is the effective dynamic viscosity. Molecular viscosity It is turbulent viscosity. It is a particle feedback force term ; Turbulent closure using SST The model is represented as: ; The recommended values for the model parameters are: , turbulent kinetic energy Turbulent dissipation rate ; S42, Particle Phase Diffusion Tracing: Based on the Feather Source Term Injected in S3 Distributed tracking of particle phases is performed within the Eulerian-Lagrange framework, with particle clouds grouped by particle size using their centroid coordinates. The evolution, the governing equations are expressed as: ; in, This represents the location of the particle's center of mass (unit: m). Particle velocity (unit: ), Local fluid velocity (unit: ), No. Relaxation time of particles (unit: s). The vector of gravitational acceleration (unit: ), This is an additional force term, including secondary effects such as Saffman lift, Magnus force, and virtual mass force (unit: N / kg). For fluid density and particle density, Fluid dynamic viscosity coefficient (unit: Pa) ), For the first Characteristic particle size of particle size group (unit: m); S43, Dynamic Reconstruction of Concentration Field: In the Grid The mass of all particles per unit volume is statistically analyzed to construct a time-varying particle concentration field. , is represented as: ; in, No. Volume of each grid cell For the particles falling into this unit 'quality, It is the first One time output point; To improve visualization smoothness, Gaussian kernel density estimation is used to generate a continuous concentration field. , is represented as: ; in, It is the smoothed concentration field. It is the total number of particles. The Euclidean distance between the grid center and the particle. The Gaussian kernel bandwidth parameter controls the smoothness; it is set to 2-5 times the particle size. S44, Spatiotemporal evolution data output: based on the constructed smooth concentration field It outputs its spatiotemporal sequence at fixed time intervals for three-dimensional visualization of plume diffusion dynamics and subsequent region division, specifically including: (1) Construction of three-dimensional concentration field output sequence: setting the time starting point With output step size Construct the overall output timing. , is represented as: ; in, This is the total number of output steps (unit: dimensionless). The smoothed concentration field (unit: ), It refers to the position of a point on a three-dimensional grid. (2) Spatial visualization format and concentration threshold setting: The output concentration field is rendered into one of the following three graphic formats to intuitively present the plume diffusion characteristics, and the output concentration range is set. Specifically, it includes: 1) Isosurface: Displays the three-dimensional distribution boundary under a fixed concentration threshold, used to assess the spatiotemporal evolution of high concentration regions; 2) Horizontal Slice: Displays a fixed water depth. The two-dimensional concentration distribution is used to analyze the near-bottom diffusion pattern; 3) Vertical Section: Along the specified cross-section ( This demonstrates the vertical structure and concentration gradient of the plume; in, For concentration visualization threshold, , .
[0025] S5 specifically includes: S51, Threshold Setting: Preset ecologically sensitive concentration thresholds, including high-risk thresholds. (unit: ), medium risk threshold (unit: and low-risk threshold (unit: The following satisfy the inequality relationship, which can be expressed as: ; S52, Automatic Region Division: Based on the Generated Time-Varying Particle Concentration Field For every moment and each spatial location Perform the following region division: (1) High-risk areas: ; (2) Medium-risk areas: ; (3) Low-risk areas: ; S53, Dynamic Risk Map Generation: Set a visualization color scheme according to risk level, where high-risk areas are marked in red, medium-risk areas in yellow, and low-risk areas in green. Output formats include: isosurface map, profile cloud map, and 3D heat map.
[0026] This invention encompasses any substitutions, modifications, equivalent methods, and solutions made within the spirit and scope of this invention. To provide the public with a thorough understanding of this invention, specific details are described in detail in the following preferred embodiments; however, those skilled in the art will fully understand the invention even without these details. Furthermore, to avoid unnecessary misunderstanding of the essence of this invention, well-known methods, processes, procedures, components, and circuits are not described in detail.
[0027] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.
Claims
1. A method for predicting the diffusion of plumes in seabed mining based on numerical simulation, characterized in that, Includes the following steps: S1, 3D Topographic Field Construction: Acquire seabed topographic measurement data of the target sea area, construct a digital topographic model, and generate the corresponding 3D mesh; S2, Environmental flow field boundary setting: Based on the constructed terrain model, historical ocean observation data and on-site sensor data are integrated to generate initial and boundary conditions for the environmental flow field, including flow velocity, flow direction and temperature-salinity structure, as the input flow field for plume simulation; S3, Particle Source Parameter Injection: Combining the emission characteristics of seabed mining equipment, a plume flow source term is set, including particle size distribution, mass concentration, emission rate and direction information, and the plume flow source term is injected into the flow field boundary to drive particle diffusion simulation; S4, Three-dimensional diffusion numerical simulation of plume flow: Based on a multi-scale CFD solver, the terrain mesh of S1, the flow field boundary of S2 and the particle source parameters of S3 are input to simulate the diffusion process of plume flow in three-dimensional space and output the spatial distribution map of particle concentration field evolution over time. S5, Dynamic Impact Domain Identification and Risk Zone Delineation: Based on the output spatial distribution map and combined with the setting of ecological sensitivity thresholds, the plume diffusion impact area is automatically delineated, high-risk, medium-risk and low-risk areas are identified, and corresponding dynamic risk maps are generated.
2. The method for predicting the diffusion of seabed mining plumes based on numerical simulation according to claim 1, characterized in that, S1 includes: S11, Acquisition of seabed topographic measurement data: Collect seabed topographic point cloud data of the target sea area through multibeam echo sounder, side-scan sonar or lidar equipment to obtain a discrete coordinate point set; S12, Terrain data preprocessing: Based on the acquired discrete coordinate point set, statistical screening of water depth data is performed to remove outliers. Spatial interpolation is then performed on the removed point set to construct a water depth function. S13, Construction of digital terrain model: Based on the Kriging interpolation function, a digital elevation model in the form of a regular network is constructed; S14, 3D Network Generation: The digital elevation model is divided into three dimensions, and the Delaunay triangulation algorithm is used to generate an unstructured tetrahedral mesh set.
3. The method for predicting the diffusion of seabed mining plumes based on numerical simulation according to claim 2, characterized in that, S2 includes: S21, Multi-source ocean data fusion: acquire historical ocean observation datasets and real-time datasets from on-site sensors, fuse the above data based on a four-dimensional variational assimilation algorithm, and construct the optimal initial state field; S22, Initialization of 3D environmental flow field: Map the optimal state field to a 3D unstructured mesh to construct the initial physical field; S23, Dynamic boundary condition setting: Based on the hydrodynamic characteristics of the seabed mining area, appropriate boundary conditions are applied to different boundary types of the numerical simulation domain. S24, Temperature and Salinity Boundary Constraint Setting: A relaxation forcing term is introduced at the open boundary to control the boundary temperature and salinity to approach the observed values.
4. The method for predicting the diffusion of seabed mining plumes based on numerical simulation according to claim 3, characterized in that, S22 includes: S221 maps the optimal state field to a three-dimensional unstructured network to construct a three-dimensional velocity field; S222 maps the optimal state field to a three-dimensional unstructured network to construct a three-dimensional temperature field and a three-dimensional salinity field.
5. The method for predicting the diffusion of seabed mining plumes based on numerical simulation according to claim 4, characterized in that, S23 includes: S231, Lateral Open Boundary Treatment: Radiation boundary conditions are used to treat the lateral boundaries of the numerical simulation region; S232, Sea Surface Boundary Modeling: Apply wind stress boundary conditions at the top edge of the simulation domain to simulate the influence of surface wind on the underlying fluid. The magnitude of wind stress is determined by air density, wind speed, and drag coefficient, while also considering turbulent viscosity effects. S233, Seabed Boundary Modeling: At the bottom edge of the simulation domain, the velocity distribution is approximated by a logarithmic wall function. Seabed roughness and friction velocity factors are introduced to reproduce the velocity profile characteristics of the plume near the seabed.
6. The method for predicting the diffusion of seabed mining plumes based on numerical simulation according to claim 5, characterized in that, S3 includes: S31, Emission Characteristic Parameterization: Based on the operating parameters of the target mining equipment, determine the emission attributes of the plume source term, including emission rate and emission direction vector; S32, Particulate matter size distribution modeling: The Rosin-Rammler distribution function is used to model the size distribution of multi-component particles and divide the particle size intervals; S33, Source Term Spatial Injection: Define the time-varying emission point locations within a three-dimensional mesh domain to construct a feather source term function; S34, Multiphase Flow Coupling Drive: Input the source term function into the Euler-Lagrange multiphase flow model.
7. The method for predicting the diffusion of seabed mining plumes based on numerical simulation according to claim 6, characterized in that, S31 includes: S311, Emission Rate Determination: Calculate the mass emission rate of the plume source based on the structural parameters and operating status of the target mining equipment; S312, Emission Direction Vector Calculation: Combining the actual attitude information of the target mining equipment, including pitch and yaw angles, the three-dimensional direction of particulate matter emission is determined, and the attitude angles are converted into direction vectors using trigonometric function relationships.
8. The method for predicting the diffusion of seabed mining plumes based on numerical simulation according to claim 7, characterized in that, S4 includes: S41, Solving the multi-scale CFD governing equations: Based on the three-dimensional terrain mesh generated in S1 and the environmental flow field boundary conditions set in S2, solve the unsteady Reynolds-averaged Navier-Stokes equations. S42, Particle Phase Diffusion Tracking: Based on the source term injected by S3, the evolution of the center and velocity of a single particle is tracked within the Eulerian-Lagrange framework; S43, Dynamic Reconstruction of Concentration Field: The mass of all particles per unit volume is statistically analyzed on the grid to construct a particle concentration field; S44, Spatiotemporal evolution data output: Outputs a three-dimensional concentration field sequence, with output formats including isosurface plots and profile cloud maps.
9. The method for predicting the diffusion of seabed mining plumes based on numerical simulation according to claim 8, characterized in that, S5 includes: S51, Threshold setting: Preset ecological sensitivity thresholds, including high-risk threshold, medium-risk threshold and low-risk threshold; S52, based on the output spatial distribution map and combined with the ecological sensitivity threshold setting, automatically divides the area affected by the plume diffusion and identifies high-risk, medium-risk and low-risk areas. S53. Based on the identified areas affected by the plume diffusion, a corresponding dynamic risk map is generated.
10. The method for predicting the diffusion of seabed mining plumes based on numerical simulation according to claim 9, characterized in that, S53 includes: S531, set a visual color scheme according to risk level; S532 generates dynamic risk maps, with output formats including: isosurface maps, profile cloud maps, and 3D heat maps.