A deep-sea three-dimensional environmental deduction and simulation analysis system based on big data

Abstract

The present invention discloses a deep - sea three - dimensional environment deduction and simulation analysis system based on big data, which relates to the technical field of deep - sea environment simulation. It includes a visualization service desk, and the visualization service desk is communicatively connected to a data acquisition and integration module, a deep - sea three - dimensional scene modeling module, a rendering optimization module, a deep - sea environment analysis module, an ocean environment deduction and simulation module, a three - dimensional visualization display module, and a deep - sea deduction typical scene roaming module. Among them, the modules are electrically connected. The present invention constructs a highly realistic deep - sea three - dimensional environment model by integrating multi - source and multi - dimensional deep - sea data, and uses big data processing technology to be able to real - time simulate the change trend of the deep - sea environment, reflect the dynamic changes of the deep - sea environment, and users can intuitively experience the deep - sea landscape through virtual roaming, significantly improving the efficiency and safety of deep - sea resource development, and reducing the development costs and risks caused by environmental uncertainties.

Description

Technical Field

[0001] The present invention relates to the technical field of deep - sea environment simulation, and particularly to a deep - sea three - dimensional environment deduction simulation analysis system based on big data. Background Technique

[0002] With the increasing demand for deep - sea resource development by humans, especially in the aspects of energy, mineral resources, marine biology, and ecosystem research, in order to better understand the deep - sea environment and conduct precise deep - sea operations, a deep - sea three - dimensional environment deduction simulation analysis system has become an essential tool. Through this system, the changes in the deep - sea environment can be monitored and analyzed in real time, providing technical support for deep - sea development and scientific research.

[0003] In deep - sea three - dimensional simulation, due to the complexity of details such as seabed topography and environment, it is difficult to ensure both high precision and real - time performance during rendering, which easily leads to problems such as insufficiently realistic display effects or slow response, and further results in a large difference between the simulation results and the actual situation. Therefore, how to improve the accuracy of deep - sea three - dimensional environment deduction simulation, enhance the authenticity of the display effect, and ensure the co - existence of accuracy and authenticity is a problem to be solved. For this reason, a deep - sea three - dimensional environment deduction simulation analysis system based on big data is proposed. Summary of the Invention

[0004] The purpose of the present invention is to provide a deep - sea three - dimensional environment deduction simulation analysis system based on big data to solve the problems raised in the above - mentioned background technique.

[0005] To solve the above - mentioned technical problems, the technical solution adopted by the present invention is:

[0006] A deep - sea three - dimensional environment deduction simulation analysis system based on big data, including a visualization service desk, which is communicatively connected to a data acquisition and integration module, a deep - sea three - dimensional scene modeling module, a rendering optimization module, a deep - sea environment analysis module, an ocean environment deduction simulation module, a three - dimensional visualization display module, and a deep - sea deduction typical scene roaming module. Among them, the modules are electrically connected to each other;

[0007] The data acquisition and integration module obtains deep - sea environment data and ship AIS data based on multiple data sources (various sensors, databases), and integrates various types of data to form a multi - dimensional deep - sea environment data set, providing a basis for subsequent analysis and deduction;

[0008] The deep - sea three - dimensional scene modeling module constructs a three - dimensional model of the deep - sea environment including seabed topography, water body, biological, and equipment elements based on the data of the deep - sea environment data set, and presents a realistic deep - sea environment through high - precision modeling;

[0009] The rendering optimization module performs real - time rendering on the three - dimensional model of the deep - sea environment based on graphics rendering technology;

[0010] The deep - sea environment analysis module further analyzes the deep - sea environment data based on machine learning algorithms to identify the changing trends in the deep - sea environment;

[0011] The marine environment deduction and simulation module conducts dynamic deduction and simulation of the deep - sea environment based on the rendered and optimized three - dimensional model of the deep - sea environment and real - time ship AIS data, combines the deep - sea environment analysis results, and predicts the future state;

[0012] The three - dimensional visualization display module is used to realize the dynamic simulation of ship navigation in a virtual scene, including the display of the movement trajectories of ships and submersibles, and dynamically updates environmental factors such as wave state, wind speed, and ocean current to enhance the authenticity of the display effect;

[0013] The deep - sea deduction typical scene roaming module is used to set locations for scene dynamic roaming, display the deduction results and mark key locations, facilitating users to quickly locate and analyze, providing a responsive roaming experience, and allowing users to independently adjust the roaming speed.

[0014] A further improvement of the technical solution of the present invention lies in that: the data acquisition and integration module specifically includes:

[0015] Identify various types of data sources of the deep - sea environment, including but not limited to deep - sea sensor networks, databases, and ship AIS data sources, and obtain deep - sea environment data and ship AIS data from various types of data sources through corresponding interfaces and protocols;

[0016] Perform pre - processing operations of data cleaning, calibration, and standardization on the obtained deep - sea environment data and ship AIS data. Among them, identify and process missing values, outliers, or duplicate data through data cleaning, perform denoising and smoothing processing on the data to improve data quality, and calibrate and standardize sensor data to ensure data consistency and accuracy. Convert AIS data from the original message format to a digital format that is easy to analyze and process;

[0017] Match the pre - processed deep - sea environment data and ship AIS data according to the time stamp and geographical location information, integrate data from different sources, form a deep - sea environment data set containing multiple dimensions, and store the deep - sea environment data set in the corresponding data warehouse for subsequent query and analysis.

[0018] A further improvement of the technical solution of the present invention lies in that: the deep - sea three - dimensional scene modeling module specifically includes:

[0019] Extract the deep - sea environment data set from the data warehouse and classify the data in the deep - sea environment data set, including seabed terrain data, water body data, biological data, and equipment data;

[0020] For seabed terrain modeling, terrain data with different resolutions (global terrain data and local high-precision terrain data) are fused to ensure the global nature and local details of the terrain model. A terrain modeling tool (World Machine) is used to construct the seabed terrain, and the terrain data is imported into the modeling software. A terrain mesh is generated through a heightmap, and then the generated terrain mesh is optimized to adjust the level of detail of the terrain.

[0021] For water body modeling, based on the physical properties of the water body (refractive index, reflectivity, scattering coefficient, etc.), a physical model of the water body is constructed, and hydrodynamic simulation technology is used to simulate the dynamic effects of water flow, waves, and ocean currents.

[0022] For biological modeling, according to the biological distribution data, a three-dimensional model library of marine organisms is constructed, including fish, plankton, and corals, etc. A modeling software (3ds Max) is used to create biological models, and then the biological models are placed in the corresponding positions to simulate their behavior patterns (swimming, foraging, reproduction, etc.), and the particle swarm optimization algorithm is combined to simulate the group behavior and movement trajectories of organisms.

[0023] For equipment modeling, a three-dimensional model library of deep-sea equipment is constructed, including scientific research ships, submersibles, and sensors, etc. According to the equipment position and data, a three-dimensional model of the deep-sea equipment is created, and the equipment model is placed in the corresponding position to simulate its dynamic behavior (navigation, diving, data collection, etc.).

[0024] The seabed terrain, water body, biological, and equipment elements are integrated into a unified three-dimensional scene to ensure the spatial relationship and time synchronization between the various elements, and the entire scene is optimized. The LOD technology is used to dynamically adjust the level of detail of the model to ensure good rendering performance at different viewing distances.

[0025] A further improvement in the technical solution of the present invention lies in that: the rendering optimization module specifically includes:

[0026] Configure Unreal Engine as the rendering engine for deep-sea three-dimensional environment deduction and simulation analysis, and initialize the rendering engine. Configure the basic parameters of the rendering engine including rendering resolution, anti-aliasing settings, and shadow quality, and then load the deep-sea three-dimensional model and related resources (textures, materials, animations, etc.) into the rendering engine.

[0027] Use the decimation tool in Unreal Engine to simplify the deep - sea 3D model, reducing the number of faces and vertices to lower the computational burden during rendering. In Unreal Engine, set multiple LOD levels for the model, and dynamically adjust the level of detail of the model according to the viewing distance and perspective of the model. Use multi - level texture technology to automatically match the corresponding texture resolution according to the viewing distance;

[0028] Configure the occlusion culling system in Unreal Engine to only render objects within the view frustum, ignoring invisible objects, reducing unnecessary rendering calculations. Enable geometry occlusion detection to cull parts occluded by other objects, further reducing the rendering load;

[0029] Use the PBR material system in Unreal Engine to optimize the reflection, refraction, and scattering parameters of the material. Combine texture compression technology (DXT) to reduce the storage and transmission overhead of texture data, implement streaming loading of textures, load texture data on demand, and avoid memory bottlenecks caused by loading too many textures at once;

[0030] Enable screen - space global illumination to reduce computational overhead and optimize the sampling of global illumination;

[0031] Use computational fluid dynamics (CFD) technology to simulate the dynamic effects of water flow, waves, and ocean currents. At the same time, reduce the amount of calculation by simplifying the model, use GPU acceleration technology (CUDA) to calculate the dynamic changes of the water body in real - time, and apply ray - tracing technology to enhance the reflection and refraction effects of the water body. At the same time, reduce the computational overhead through screen - space reflection (SSR) technology;

[0032] Cache the animations of organisms and devices to avoid repeated calculations. Use animation reuse technology to reduce the storage and transmission overhead of animation data, and use skeletal animation technology to achieve the dynamic behavior of organisms and devices, optimizing bone weights and animation frame rates to reduce the amount of calculation;

[0033] Decompose the rendering task into multiple subtasks, use multi - threading technology to distribute the rendering tasks to multiple CPU cores for parallel processing, use GPU - accelerated rendering to utilize the powerful parallel computing ability of the GPU to improve the rendering speed, and monitor the rendering performance in real - time, including frame rate, CPU and GPU usage, and memory occupancy. Use a performance analysis tool (NVIDIA Nsight) to locate performance bottlenecks.

[0034] A further improvement of the technical solution of the present invention lies in that: the deep - sea environment analysis module specifically includes:

[0035] Extract the integrated deep - sea environmental data including seabed terrain data, water body data, biological data, and equipment data, perform standardization processing on the data to eliminate the dimensional and numerical range differences between different data sources, and extract features related to the analysis target. Among them, extract terrain features from the seabed terrain data, including terrain slope, depth change rate, and terrain roughness; extract hydrological features from the water body data, including water temperature gradient, salinity change, and water flow velocity field; extract biological features from the biological data, including species density, distribution range, and behavior pattern; extract equipment status features from the equipment data, including equipment location, movement trajectory, and data acquisition frequency;

[0036] Analyze historical deep - sea environmental data, classify the data into different categories and label them. Among them, the seabed terrain is divided into stable areas, erosion areas, and deposition areas; the water body characteristics are divided into thermocline, halocline, and vortex areas; the biological distribution is divided into high - density areas, low - density areas, and migration paths; the equipment status is divided into normal operation, fault status, and data anomaly;

[0037] Use historical deep - sea environmental data combined with machine - learning algorithms to train a machine - learning model, optimize the model parameters through methods such as cross - validation, and evaluate the model performance to ensure the accuracy and generalization ability of the model for analyzing the change trend of deep - sea environmental data;

[0038] Use the trained machine - learning model to predict and analyze deep - sea environmental data, and identify the change trends in the deep - sea environment, including seabed terrain evolution trends, water body change trends, biological distribution and behavior trends, and equipment operation trends.

[0039] A further improvement of the technical solution of the present invention lies in: The calculation process of the seabed terrain evolution trend is as follows:

[0040] Collect seabed terrain data and , respectively represent the terrain depths of the \(i\) - th measurement point at times \(t\) and \(t - 1\), and determine the time interval , select the attenuation factor ;

[0041] For each measurement point \(i\), calculate its depth change rate within the time interval , and apply an exponential decay function to the depth change rate of each measurement point to smooth the mutation of depth change and reduce the influence of extreme values on the trend;

[0042] Sum up the adjusted depth change rates of all measurement points to obtain the seabed terrain evolution trend of the entire region , if >0, it indicates that the overall terrain depth increases and there is sedimentation. If < 0 indicates a decrease in the overall terrain depth, indicating erosion;

[0043] The calculation process of the water body change trend is as follows:

[0044] Collect water body parameter data and , which respectively represent the values of the j-th water body parameter at time t and t - 1, and determine the time interval , select the adjustment factor ;

[0045] For each water body parameter j, calculate its water body parameter change rate within the time interval , and apply a non-linear adjustment function to the change rate of each water body parameter to enhance significant changes and reduce the influence of minor changes;

[0046] Sum up the adjusted change rates of all water body parameters to obtain the water body change trend of the entire region , if > 0, it indicates an increase in the overall water body parameters, such as an increase in water temperature, an increase in salinity, etc. If < 0, it indicates a decrease in the overall water body parameters, such as a decrease in water temperature, a decrease in water flow velocity, etc.;

[0047] The calculation process of the biological distribution and behavior trend is as follows:

[0048] Collect biological characteristic data and , which respectively represent the values of the k-th biological characteristic at time t and t - 1, and determine the time interval , select the adjustment factor ;

[0049] For each biological characteristic k (from 1 to p), calculate its biological characteristic change rate within the time interval , and apply a logarithmic function to the change rate of each biological characteristic to smooth the change of biological characteristics and avoid the influence of extreme values on the trend;

[0050] Sum up the adjusted change rates of all biological characteristics to obtain the biological distribution and behavior trend of the entire region , if > 0, it indicates an increase in the overall biological characteristics, such as an increase in species density, an expansion of the distribution range, etc. If < 0, it indicates a decrease in the overall biological characteristics, such as a decrease in species density, a change in behavior patterns, etc.;

[0051] The calculation process of the equipment operation trend is as follows:

[0052] Collect equipment status parameter data and , representing the values of the l-th device status parameter at time t and t-1 respectively, and determining the time interval , and selecting an adjustment factor ;

[0053] For each device status parameter l, calculate the change rate of the device status parameter within the time interval , and apply the Sigmoid function to the change rate of each device status parameter to smooth the change of the device status and avoid the influence of extreme values on the trend;

[0054] Sum up the adjusted change rates of all device status parameters to obtain the operation trend of the devices in the entire area , >0 indicates that the overall device status parameter increases, and there is a situation such as an increase in the data acquisition frequency. If <0, it indicates that the overall device status parameter decreases, and there is a situation such as a change in the device position.

[0055] A further improvement of the technical solution of the present invention lies in that: the marine environment deduction and simulation module specifically includes:

[0056] Using the three-dimensional model of the deep-sea environment generated by the rendering optimization module, including elements such as seabed topography, water body, organisms, and devices, ensuring the real-time and high-precision rendering of the model, providing a realistic visual effect for the simulation, and accessing real-time ship AIS data, including ship position, speed, and heading information;

[0057] Combining the analysis results of the deep-sea environment analysis module including the seabed topography evolution trend, water body change trend, biological distribution and behavior trend, and device operation trend, and integrating the historical deep-sea environment data with the analysis results;

[0058] According to the three-dimensional model and real-time data at the current time point, initialize the state of the deep-sea environment, including the initial state of the seabed topography, the initial conditions of the water body, the initial distribution and behavior patterns of organisms, and the initial positions and states of devices;

[0059] Implement physical process simulation, including water body dynamics simulation, biological behavior simulation, and device dynamics simulation. Among them, use the hydrodynamic model to simulate the dynamic changes of water flow, waves, and ocean currents, combine the water body change trend to predict the future temperature, salinity, and water velocity field of the water body, use the behavior model (swarm behavior model based on particle swarm optimization) to simulate the migration, foraging, and reproduction behaviors of organisms, combine the biological distribution and behavior trend to predict the future distribution and behavior patterns of organisms, use the kinematics and dynamics models to simulate the movement trajectories of ships and submersibles, and combine the device operation trend to predict the future states and potential fault points of devices;

[0060] Adopt the time-stepping method to gradually update the state of the deep-sea environment. And within each time step, update the state of the 3D model according to the simulation results of physical processes to ensure the accuracy and real-time nature of the simulation results;

[0061] Based on the results of dynamic deduction, predict the future changes in the deep-sea environment, realize the long-term evolution prediction of the seabed topography, the seasonal changes and abnormal fluctuations prediction of water bodies, the distribution and behavior changes prediction of organisms, and the operating state prediction of equipment, and generate a scientific prediction report, including the visual display and written description of the prediction results, and provide a scientific explanation of the prediction results to help users understand the future changes in the deep-sea environment.

[0062] A further improvement of the technical solution of the present invention lies in that: the process of updating the state of the 3D model is as follows:

[0063] Define the time step, set it to 1 hour, and gradually advance the simulation time according to the set time step;

[0064] Within each time step, update the state of the 3D model according to the simulation results of physical processes, and update the seabed topography, water body state, biological population distribution, and equipment state in the 3D model according to the simulation results of physical and ecological processes;

[0065] Take the updated state as the initial condition for the next time step for iterative calculation. After the simulation ends, output the required result data, including the changes in the seabed topography, the seasonal changes and abnormal fluctuations of water bodies, the distribution and behavior changes of organisms, and the operating state of equipment, and then perform visual processing on the result data to intuitively analyze the changing trend of the deep-sea environment.

[0066] A further improvement of the technical solution of the present invention lies in that: the 3D visualization display module specifically includes:

[0067] Load the 3D model of the deep-sea environment generated by the rendering optimization module, import the loaded 3D model into the virtual scene, set the initial state of the scene, including lighting, shadows, environmental parameters, etc., and at the same time configure the basic parameters of the rendering engine to ensure that the rendering engine can efficiently process complex deep-sea environment scenes;

[0068] Access real-time ship AIS data, including ship position, speed, and heading information, synchronize the AIS data to the ship and submersible models in the 3D model in real time, and use kinematic and dynamic models, combined with the equipment operation trend, to simulate the movement trajectories of ships and submersibles, and dynamically display the real-time movements of ships and submersibles in the 3D scene, including the navigation path and the diving trajectory. According to the water body dynamics simulation results, dynamically update environmental factors such as wave state, wind speed, and ocean current, and render the wave, water flow, and wind field effects in the 3D scene in real time to enhance the authenticity of the display effect;

[0069] Provide a multi-scenario resource switching function, enabling users to observe the deep-sea environment from different perspectives, including global view, local view, and specific area view. Users can switch scenarios according to their needs and provide a user interface that allows users to perform scenario switching and navigation through mouse, keyboard, or touch screen operations.

[0070] A further improvement of the technical solution of the present invention lies in: the deep-sea deduction typical scenario roaming module specifically includes:

[0071] According to the user's needs, configure the relevant parameters of the roaming module, including roaming speed, viewing angle range, and scene accuracy, and read the deep-sea scene data required for deduction from the database, including seabed terrain, water body state, and biological distribution, etc., to ensure that the data is accurate, complete, and compatible with the model;

[0072] Automatically navigate to the target location according to the location input by the user and start dynamic roaming. During the roaming process, dynamically update the rendering effect of the scene according to the user's perspective and position, and provide real-time feedback on the user's operations to ensure the smoothness and real-time nature of the roaming process. At the same time, provide a user interface that allows users to independently adjust the roaming speed, supporting multiple speed modes including fast, medium, and slow, and users can select the corresponding roaming speed according to their needs;

[0073] Dynamically display the deduction results in a three-dimensional scene, highlighting key information using different colors, icons, or animation effects for easy user identification, and provide a marking function that allows users to mark key locations in the three-dimensional scene. The marking includes text annotations, icons, or color highlighting. At the same time, support users to quickly navigate to key locations through marking, and provide a query function. Users can input the location name or marking number to locate to the target position.

[0074] Due to the adoption of the above technical solution, the technical progress achieved by the present invention compared with the prior art is:

[0075] The present invention provides a deep-sea three-dimensional environment deduction and simulation analysis system based on big data. By integrating multi-source and multi-dimensional deep-sea data, a highly realistic deep-sea three-dimensional environment model is constructed. Using big data processing technology, it can real-time simulate the change trend of the deep-sea environment, reflect the dynamic changes of the deep-sea environment, and users can intuitively experience the deep-sea landscape through virtual roaming, significantly improving the efficiency and safety of deep-sea resource development, and reducing the development costs and risks caused by environmental uncertainties.

[0076] The present invention provides a deep - sea three - dimensional environment deduction simulation analysis system based on big data. By dynamically deducing the change trend of the deep - sea environment, it can predict the changes in biological distribution and behavior patterns, helping researchers and managers better understand the dynamic balance of the marine ecosystem. In addition, through the three - dimensional visualization display module, the interaction between organisms and the environment is intuitively presented, which can effectively reduce the negative impact of human activities on the marine ecosystem and promote the sustainable utilization of marine resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0077] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments recorded in the present invention. For those of ordinary skill in the art, other drawings can also be obtained based on these drawings.

[0078] Figure 1 It is a schematic diagram of the system function modules of the present invention;

[0079] Figure 2 It is a schematic diagram of the working process of the deep - sea environment analysis module of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0080] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0081] Embodiment 1, as Figure 1 、 Figure 2 shown, the present invention provides a deep - sea three - dimensional environment deduction simulation analysis system based on big data, including a visualization service desk, which is communicatively connected to a data acquisition and integration module, a deep - sea three - dimensional scene modeling module, a rendering optimization module, a deep - sea environment analysis module, a marine environment deduction simulation module, a three - dimensional visualization display module, and a deep - sea deduction typical scene roaming module. Among them, the modules are electrically connected to each other;

[0082] Data acquisition and integration module, which acquires deep - sea environment data and ship AIS data based on multiple data sources (various sensors, databases), integrates various types of data to form a multi - dimensional deep - sea environment data set, provides a basis for subsequent analysis and deduction, identifies various types of data sources for the deep - sea environment, including but not limited to deep - sea sensor networks, databases, and ship AIS data sources, and obtains deep - sea environment data and ship AIS data from various types of data sources through corresponding interfaces and protocols. Among them, ship AIS data includes AIS base station data and satellite AIS data. AIS base station data is to receive AIS messages sent by ships using shore - based AIS equipment, including ship position, speed, and heading information. Satellite AIS data is to receive ship signals through high - sensitivity AIS receivers carried by low - orbit satellites. Perform pre - processing operations such as data cleaning, calibration, and standardization on the acquired deep - sea environment data and ship AIS data. Among them, identify and process missing values, outliers, or duplicate data through data cleaning, perform denoising and smoothing processing on the data to improve data quality, and calibrate and standardize sensor data to ensure data consistency and accuracy. Convert AIS data from the original message format to a digital format that is easy to analyze and process. Match the pre - processed deep - sea environment data and ship AIS data according to timestamp and geographical location information, integrate data from different sources to form a deep - sea environment data set containing multiple dimensions, and store the deep - sea environment data set in the corresponding data warehouse for subsequent query and analysis;

[0083] Deep - sea three - dimensional scene modeling module, based on the data of the deep - sea environment dataset, constructs a three - dimensional model of the deep - sea environment including seabed topography, water body, organisms and equipment elements. Through high - precision modeling, a realistic deep - sea environment is presented. Extract the deep - sea environment dataset from the data warehouse and classify the data in the deep - sea environment dataset, including seabed topography data, water body data, biological data and equipment data. Among them, seabed topography data includes global topography data, multibeam sounding data, seafloor geomorphology data, etc.; water body data includes data such as water temperature, salinity, water flow velocity, water quality, etc.; biological data includes the distribution, species, behavior patterns of marine organisms, etc.; equipment data includes the position and status information of equipment such as scientific research vessels, submersibles, sensors, etc. For seabed topography modeling, fuse topographic data with different resolutions (global topography data and local high - precision topography data) to ensure the global nature and local details of the topographic model. Use a topographic modeling tool (World Machine) to construct the seabed topography, import the topographic data into the modeling software, generate a terrain mesh through a height map, and then optimize the generated terrain mesh to adjust the level of detail of the terrain. For water body modeling, based on the physical properties of the water body (refractive index, reflectivity, scattering coefficient, etc.), construct a physical model of the water body and use hydrodynamic simulation technology to simulate the dynamic effects of water flow, waves and ocean currents. For biological modeling, according to the biological distribution data, construct a three - dimensional model library of marine organisms, including fish, plankton and corals, etc., and use modeling software (3ds Max) to create biological models, and then place the biological models in the corresponding positions to simulate their behavior patterns (swimming, foraging, reproduction, etc.), and combine the particle swarm optimization algorithm to simulate the group behavior and movement trajectories of organisms. For equipment modeling, construct a three - dimensional model library of deep - sea equipment, including scientific research vessels, submersibles and sensors, etc. According to the equipment position and data, create a three - dimensional model of the deep - sea equipment, place the equipment model in the corresponding position, and simulate its dynamic behavior (navigation, diving, data collection, etc.). Integrate the seabed topography, water body, organisms and equipment elements into a unified three - dimensional scene, ensure the spatial relationship and time synchronization between each element, and optimize the entire scene. Use LOD technology to dynamically adjust the level of detail of the model to ensure good rendering performance at different viewing distances;

[0084] Render optimization module, based on graphics rendering technology, performs real-time rendering on the three-dimensional model of the deep-sea environment to ensure rendering quality. While ensuring simulation accuracy, it improves the system's response speed and display effect, achieving a balance between accuracy and real-time performance. Configure Unreal Engine as the rendering engine for deep-sea three-dimensional environment deduction simulation analysis, and initialize the rendering engine. Configure the basic parameters of the rendering engine, including rendering resolution, anti-aliasing settings, and shadow quality. Then load the deep-sea three-dimensional model and related resources (textures, materials, animations, etc.) into the rendering engine. Use Unreal Engine's decimation tool to simplify the deep-sea three-dimensional model, reducing the number of faces and vertices to reduce the computational burden during rendering. In Unreal Engine, set multiple LOD levels for the model, and dynamically adjust the level of detail of the model according to the viewing distance and perspective of the model. Use multi-level texture technology to automatically match the corresponding texture resolution according to the viewing distance, reducing the computational amount of texture rendering. In the case of a distant or small perspective, use a low-precision model to reduce the rendering burden. When close or with a large perspective, switch to a high-precision model and synchronously use multi-level texture technology. Configure Unreal Engine's occlusion culling system to only render objects within the view frustum and ignore invisible objects, reducing unnecessary rendering calculations. Enable geometry occlusion detection to cull parts occluded by other objects, further reducing the rendering load. Use Unreal Engine's PBR material system to optimize the reflection, refraction, and scattering parameters of the materials, ensuring the realism and consistency of the materials under different lighting conditions, reducing computational complexity, and combining texture compression technology (DXT) to reduce the storage and transmission overhead of texture data, realizing the streaming loading of textures, loading texture data on demand, and avoiding memory bottlenecks caused by loading too many textures at once. Enable screen space global illumination to reduce computational overhead, optimize the sampling of global illumination to reduce noise and computational time, and use screen space ambient occlusion technology to reduce the complexity of shadow calculations. Optimize the shadow map using high-resolution shadow maps to enhance the realism and performance of shadows. Use computational fluid dynamics (CFD) technology to simulate the dynamic effects of water flow, waves, and ocean currents, while reducing the computational amount by simplifying the model. Use GPU acceleration technology (CUDA) to calculate the dynamic changes of the water body in real time, and apply ray tracing technology to enhance the reflection and refraction effects of the water body, while reducing the computational overhead through screen space reflection (SSR) technology. Cache the animations of organisms and devices to avoid repeated calculations, use animation reuse technology to reduce the storage and transmission overhead of animation data, and use skeletal animation technology to achieve the dynamic behavior of organisms and devices, optimizing the bone weights and animation frame rates to reduce the computational amount. Decompose the rendering task into multiple subtasks, utilize multi-threading technology, and distribute the rendering tasks to multiple CPU cores for parallel processing. Use GPU-accelerated rendering to utilize the powerful parallel computing ability of the GPU to improve the rendering speed.Monitor the rendering performance in real time, including frame rate, CPU and GPU usage, and memory occupancy. Use a performance analysis tool (NVIDIA Nsight) to locate performance bottlenecks;

[0085] Deep - sea environment analysis module, which further analyzes deep - sea environment data based on machine - learning algorithms to identify changing trends in the deep - sea environment;

[0086] Marine environment deduction and simulation module, based on the three - dimensional model of the deep - sea environment optimized for rendering and real - time ship AIS data, combined with the results of deep - sea environment analysis, dynamically deduce and simulate the deep - sea environment, predict future states, provide scientific predictions of deep - sea environment changes, and provide decision - making support for deep - sea resource development, ecological protection, etc.;

[0087] Three - dimensional visualization display module, used to implement the dynamic simulation of ship navigation in a virtual scene, including the display of the movement trajectories of ships and submersibles, and dynamically update environmental factors such as wave state, wind speed, and ocean current to enhance the authenticity of the display effect, and provide a multi - scene resource switching display function to facilitate users to observe the deep - sea environment from different angles;

[0088] Deep - sea deduction typical scenario roaming module, used to set locations for dynamic scenario roaming, display deduction results and mark key locations to facilitate users to quickly locate and analyze, provide a responsive roaming experience, and allow users to independently adjust the roaming speed.

[0089] Example 2, as Figure 1 、 Figure 2 shown, on the basis of Example 1, the present invention provides a technical solution: Preferably, the deep - sea environment analysis module specifically includes:

[0090] Extract the integrated deep-sea environmental data including seabed terrain data, water body data, biological data, and equipment data, perform standardization processing on the data to eliminate the dimensional and numerical range differences between different data sources, and extract the features related to the analysis target. Among them, extract terrain features from the seabed terrain data, including terrain slope, depth change rate, and terrain roughness; extract hydrological features from the water body data, including water temperature gradient, salinity change, and water flow velocity field; extract biological features from the biological data, including species density, distribution range, and behavior pattern; extract equipment status features from the equipment data, including equipment location, movement trajectory, and data collection frequency. Analyze the historical deep-sea environmental data, classify the data into different categories, and perform annotation. Among them, the seabed terrain is divided into stable areas, erosion areas, and deposition areas; the water body features are divided into thermocline, halocline, and vortex areas; the biological distribution is divided into high-density areas, low-density areas, and migration paths; the equipment status is divided into normal operation, fault status, and data anomaly. Use the historical deep-sea environmental data combined with machine learning algorithms to train a machine learning model, optimize the model parameters through methods such as cross-validation, and evaluate the model performance to ensure the accuracy and generalization ability of the model for analyzing the change trend of deep-sea environmental data. Among them, the ARIMA model can be selected to analyze the trends of water temperature, water flow velocity, etc. over time; the K-means algorithm can be selected to identify biological distribution patterns or seabed terrain features; the neural network can be selected to predict equipment failures or biological behavior patterns. Use the trained machine learning model to predict and analyze the deep-sea environmental data, and identify the change trends in the deep-sea environment, including seabed terrain evolution trends, water body change trends, biological distribution and behavior trends, and equipment operation trends. Among them, by analyzing the seabed terrain evolution trend, identify the long-term changes in the seabed terrain; by analyzing the water body change trend, identify the seasonal changes, long-term trends, and abnormal fluctuations of the water body; by analyzing the biological distribution and behavior trends, identify the distribution patterns, migration paths, and behavior change trends of organisms; through the equipment operation trend, predict the changes in the equipment operation status and potential fault points;

[0091] The calculation process of the seabed terrain evolution trend is as follows:

[0092] Collect seabed terrain data and , which respectively represent the terrain depths of the i-th measurement point at times t and t - 1, and determine the time interval , select the attenuation factor , for each measurement point i, calculate its depth change rate within the time interval , and apply an exponential decay function to the depth change rate of each measurement point to smooth the sudden changes in depth and reduce the impact of extreme values on the trend. Sum up the adjusted depth change rates of all measurement points to obtain the seabed terrain evolution trend of the entire region , if > 0 indicates an increase in the overall seabed terrain depth, indicating sedimentation. If < 0 indicates a decrease in the overall seabed terrain depth, indicating erosion;

[0093] Expression for the trend of seabed terrain evolution:

[0094] ;

[0095] In the formula, represents the trend of seabed terrain evolution at time t, represents the terrain depth of the i-th measurement point at time t, represents the terrain depth of the i-th measurement point at time t - 1, which is the time interval, represents the time difference between two measurements, is the attenuation factor used to adjust the sensitivity of depth changes. n is the total number of measurement points, is the exponential decay function used to smooth the abrupt changes in depth changes. When it indicates an increase in terrain depth, and when it indicates a decrease in terrain depth;

[0096] The calculation process of the water body change trend is as follows:

[0097] Collect water body parameter data and , which represent the values of the j-th water body parameter at times t and t - 1 respectively, and determine the time interval , select the adjustment factor . For each water body parameter j, calculate its water body parameter change rate within the time interval , and apply a non-linear adjustment function to the change rate of each water body parameter to enhance significant changes and reduce the influence of minor changes. Sum the adjusted change rates of all water body parameters to obtain the water body change trend of the entire area . If > 0, it indicates an increase in the overall water body parameters, such as an increase in water temperature and salinity. If < 0, it indicates a decrease in the overall water body parameters, such as a decrease in water temperature and a decrease in water flow velocity;

[0098] Expression for the water body change trend:

[0099] ;

[0100] In the formula, represents the water body change trend at time t, represents the value of the j-th water body parameter at time t, represents the value of the j-th water body parameter at time t - 1, is the time interval, is the adjustment factor used to control the sensitivity of the change, where m is the total number of water body parameters, is a non-linear adjustment function used to enhance the significance of the change. When it indicates an increase in the water body parameter, and when it indicates a decrease in the water body parameter;

[0101] The calculation process of the biological distribution and behavior trend is as follows:

[0102] Collect biological characteristic data and , which represent the values of the k-th biological characteristic at times t and t - 1 respectively, and determine the time interval , select the adjustment factor . For each biological characteristic k (from 1 to p), calculate its biological characteristic change rate within the time interval , and apply the logarithmic function to the change rate of each biological characteristic to smooth the change of the biological characteristic and avoid the influence of extreme values on the trend. Sum up the adjusted change rates of all biological characteristics to obtain the biological distribution and behavior trend of the entire region , if > 0, it indicates an increase in the overall biological characteristics, such as an increase in species density and an expansion of the distribution range. If < 0, it indicates a decrease in the overall biological characteristics, such as a decrease in species density and a change in the behavior pattern;

[0103] Expression of the biological distribution and behavior trend:

[0104] ;

[0105] In the formula, represents the biological distribution and behavior trend at time t, represents the value of the k-th biological characteristic at time t, represents the value of the k-th biological characteristic at time t - 1, is the time interval, is the adjustment factor used to control the sensitivity of the biological characteristic change, where p is the total number of biological characteristics, is the logarithmic function used to smooth the change of the biological characteristic. When it indicates an increase in the biological characteristic, and when it indicates a decrease in the biological characteristic;

[0106] The calculation process of the device operation trend is as follows:

[0107] Collect device status parameter data and , respectively representing the values of the l-th device status parameter at time t and t - 1, and determining the time interval , select the adjustment factor , for each device status parameter l, calculate its device status parameter change rate within the time interval , and apply the Sigmoid function to the change rate of each device status parameter to smooth the change of the device status, avoid the influence of extreme values on the trend, sum up the adjusted change rates of all device status parameters, and obtain the device operation trend of the entire area , >0 indicates that the overall device status parameter increases, and there is a situation such as an increase in the data acquisition frequency. If <0 indicates that the overall device status parameter decreases, and there is a situation such as a change in the device position;

[0108] Expression of the device operation trend:

[0109] ;

[0110] In the formula, represents the device operation trend at time t, represents the value of the l-th device status parameter at time t, represents the value of the l-th device status parameter at time t - 1, is the time interval, is the adjustment factor, used to control the sensitivity of the device status change, q is the total number of device status parameters, is the Sigmoid function, used to smooth the change of the device status. When , it indicates that the device status parameter increases. When , it indicates that the device status parameter decreases.

[0111] Example 3, as Figure 1 、 Figure 2 shown, on the basis of Examples 1 - 2, the present invention provides a technical solution: Preferably, the marine environment deduction and simulation module specifically includes:

[0112] A three-dimensional model of the deep-sea environment generated using a rendering optimization module, including elements such as seabed topography, water body, organisms, and equipment, ensuring real-time and high-precision rendering of the model, providing a realistic visual effect for the simulation, and accessing real-time ship AIS data, including ship position, speed, and heading information, combining the analysis results of a deep-sea environment analysis module including seabed topography evolution trends, water body change trends, organism distribution and behavior trends, and equipment operation trends, and integrating historical deep-sea environment data with the analysis results. Initialize the state of the deep-sea environment based on the three-dimensional model and real-time data at the current time point, including the initial state of the seabed topography, the initial conditions of the water body, the initial distribution and behavior patterns of organisms, and the initial position and state of equipment. Implement physical process simulations, including water body dynamics simulation, organism behavior simulation, and equipment dynamics simulation. Among them, use a hydrodynamic model to simulate the dynamic changes of water flow, waves, and ocean currents, combine with the water body change trends to predict the future temperature, salinity, and water velocity field of the water body. Use a behavior model (a swarm behavior model based on particle swarm optimization) to simulate the migration, foraging, and reproduction behaviors of organisms, combine with the organism distribution and behavior trends to predict the future distribution and behavior patterns of organisms. Use kinematic and dynamic models to simulate the movement trajectories of ships and submersibles, combine with the equipment operation trends to predict the future state and potential fault points of the equipment. Adopt a time-stepping method to gradually update the state of the deep-sea environment, and within each time step, update the state of the three-dimensional model according to the simulation results of the physical processes to ensure the accuracy and real-time nature of the simulation results. Based on the results of dynamic deduction, predict the change trends of the deep-sea environment in the future, realize long-term evolution prediction of the seabed topography, seasonal change and abnormal fluctuation prediction of the water body, distribution and behavior change prediction of organisms, and operation state prediction of equipment, and generate a scientific prediction report, including visual display and text description of the prediction results, provide a scientific explanation of the prediction results to help users understand the future change trends of the deep-sea environment. Among them, based on the initial state and evolution trends of the seabed topography, analyze the scouring and sedimentation effects of water flow on the seabed topography and predict the long-term change trends of the seabed topography. Based on the initial conditions and change trends of the water body, use a hydrodynamic model to simulate the seasonal changes of the water body, combine with meteorological data and ocean observation data to predict the abnormal fluctuations of the water body. Based on the initial distribution and behavior patterns of organisms, as well as the distribution and behavior trends, use an ecological model to predict the distribution range, quantity, and behavior changes of organism populations and predict the future distribution and behavior changes of organisms. Based on the initial position and state of equipment, as well as the operation trends, analyze the force conditions of the equipment in the deep-sea environment and predict its operation state and lifespan;

[0113] In addition, the process of updating the state of the three-dimensional model is as follows:

[0114] Define the time step, set it to 1 hour, and gradually advance the simulation time according to the set time step. Within each time step, update the state of the 3D model based on the simulation results of the physical process, and update the seabed topography, water body state, biological population distribution, and equipment state in the 3D model according to the simulation results of the physical and ecological processes. Use the updated state as the initial condition for the next time step to perform iterative calculations. After the simulation ends, output the required result data, including the changes in the seabed topography, seasonal changes and abnormal fluctuations of the water body, the distribution and behavior changes of organisms, and the operating state of the equipment. Then, perform visualization processing on the result data to intuitively analyze the change trend of the deep-sea environment;

[0115] The 3D visualization display module specifically includes:

[0116] Load the 3D model of the deep-sea environment generated by the rendering optimization module, import the loaded 3D model into the virtual scene, set the initial state of the scene, including lighting, shadows, environmental parameters, etc., and at the same time configure the basic parameters of the rendering engine to ensure that the rendering engine can efficiently process complex deep-sea environment scenes. Access real-time ship AIS data, including ship position, speed, and heading information, synchronize the AIS data to the ship and submersible models in the 3D model in real time, and use kinematic and dynamic models to simulate the movement trajectories of the ship and submersible in combination with the equipment operation trend. Dynamically display the real-time movement of the ship and submersible in the 3D scene, including the navigation path and the diving trajectory. According to the simulation results of water body dynamics, dynamically update environmental factors such as wave state, wind speed, and ocean current, and real-time render wave, water flow, and wind field effects in the 3D scene to enhance the authenticity of the display effect. Provide a multi-scene resource switching function, allowing users to observe the deep-sea environment from different angles, including global views, local views, and specific area views. Users can switch scenes according to their needs and provide a user interaction interface that allows users to perform scene switching and navigation operations through mouse, keyboard, or touch screen operations;

[0117] The deep-sea deduction typical scene roaming module specifically includes:

[0118] Configure the relevant parameters of the roaming module according to the user's needs, including roaming speed, viewing angle range, and scene accuracy. Read the deep-sea scene data required for deduction from the database, including seabed topography, water body state, and biological distribution, etc., to ensure that the data is accurate, complete, and compatible with the model. Automatically navigate to the target location according to the location input by the user and start dynamic roaming. During the roaming process, dynamically update the rendering effect of the scene according to the user's viewing angle and position, and provide real-time feedback on the user's operations to ensure the smoothness and real-time nature of the roaming process. At the same time, provide a user interface that allows the user to independently adjust the roaming speed, supporting multiple speed modes including fast, medium, and slow. The user can select the corresponding roaming speed according to their needs. Dynamically display the deduction results in a 3D scene, highlighting key information using different colors, icons, or animation effects for easy user identification, and provide a marking function that allows the user to mark key locations in the 3D scene. The markings include text annotations, icons, or color highlights. At the same time, support the user to quickly navigate to the key locations through the markings. Provide a query function where the user can input the location name or marking number to locate to the target position.

[0119] As described above, it is only the specific implementation manner of this application, but the protection scope of this application is not limited thereto. Any person skilled in the art within the technical scope disclosed by this application can easily think of changes or substitutions, which should all be covered within the protection scope of this application. Therefore, the protection scope of this application should be subject to the protection scope of the claims.

Claims

1. A deep - sea three - dimensional environment deduction and simulation analysis system based on big data, including a visualization service desk, characterized in that: The visualization service desk is communicatively connected to a data acquisition and integration module, a deep-sea three-dimensional scene modeling module, a rendering optimization module, a deep-sea environment analysis module, a marine environment deduction and simulation module, a three-dimensional visualization display module, and a deep-sea deduction typical scene roaming module. Among them, the modules are electrically connected to each other; The data acquisition and integration module acquires deep-sea environment data and ship AIS data based on multiple data sources, and integrates various types of data to form a multi-dimensional deep-sea environment data set; The deep-sea three-dimensional scene modeling module constructs a three-dimensional model of the deep-sea environment including seabed topography, water body, organisms, and equipment elements based on the data in the deep-sea environment data set; The rendering optimization module performs real-time rendering on the three-dimensional model of the deep-sea environment based on graphics rendering technology; The deep-sea environment analysis module further analyzes the deep-sea environment data based on machine learning algorithms to identify the changing trends in the deep-sea environment. The deep-sea environment analysis module specifically includes: Extract the integrated deep-sea environment data including seabed topography data, water body data, organism data, and equipment data, perform standardization processing on the data, and extract features related to the analysis target; Analyze historical deep-sea environment data, classify the data into different categories, and perform annotation. Among them, the seabed topography is divided into stable areas, erosion areas, and deposition areas, the water body characteristics are divided into thermocline, halocline, and vortex areas, the biological distribution is divided into high-density areas, low-density areas, and migration paths, and the equipment status is divided into normal operation, fault status, and data anomaly; Use historical deep-sea environment data combined with machine learning algorithms to train a machine learning model and evaluate the model performance to analyze the changing trends of deep-sea environment data; Use the trained machine learning model to predict and analyze the deep-sea environment data, and identify the changing trends in the deep-sea environment, including seabed topography evolution trends, water body change trends, biological distribution and behavior trends, and equipment operation trends. The calculation process of the seabed topography evolution trend is: Collect seabed terrain data and , respectively represent the terrain depths of the i-th measurement point at time t and t - 1, and determine the time interval , select the attenuation factor; For each measurement point i, calculate its depth change rate within the time interval and apply an exponential decay function to the depth change rate of each measurement point; Sum the adjusted depth change rates of all measurement points to obtain the seabed terrain evolution trend of the entire area , if > 0, it indicates that the overall terrain depth increases. If < 0, it indicates that the overall terrain depth decreases; The calculation process of the water body change trend is: Collect water body parameter data and , representing the values of the j-th water body parameter at times t and t-1 respectively, and determining the time interval , select the adjustment factor ; For each water body parameter j, calculate the rate of change of the water body parameter within the time interval and apply a non-linear adjustment function to the rate of change of each water body parameter; Sum the adjusted change rates of all water body parameters to obtain the water body change trend of the entire area , if > 0, it indicates an increase in the overall water body parameters. If < 0, it indicates a decrease in the overall water body parameters; The calculation process of the biological distribution and behavior trend is: Collect biometric data and , representing the values of the k-th biometric feature at times t and t-1 respectively, and determining the time interval , select the adjustment factor ; For each biometric feature k, calculate its biometric change rate within the time interval and apply a logarithmic function to the change rate of each biometric feature; Sum the adjusted change rates of all biometric features to obtain the biological distribution and behavior trends of the entire area , if > 0, it indicates an increase in the overall biometric features. If < 0, it indicates a decrease in the overall biometric features; The calculation process of the equipment operation trend is: Collect device status parameter data and , representing the values of the l-th device status parameter at time t and t-1 respectively, and determining the time interval , select the adjustment factor ; For each device status parameter l, calculate the change rate of the device status parameter within the time interval and apply the Sigmoid function to the change rate of each device status parameter; Sum the adjusted change rates of all device status parameters to obtain the device operation trend of the entire area , > 0 indicates an increase in the overall device status parameters. If < 0, it indicates a decrease in the overall device status parameters; The marine environment deduction and simulation module performs dynamic deduction and simulation on the deep-sea environment based on the rendered and optimized three-dimensional model of the deep-sea environment and real-time ship AIS data, combined with the deep-sea environment analysis results; The three-dimensional visualization display module is used to realize the dynamic simulation of ship navigation in a virtual scene, including the display of the movement trajectories of ships and submersibles; The deep-sea deduction typical scene roaming module is used to set locations for scene dynamic roaming, display the deduction results, and mark key locations.

2. The three-dimensional deep-sea environment deduction and simulation analysis system based on big data according to claim 1, wherein: The data acquisition and integration module specifically includes: Identify various types of data sources for the deep-sea environment, including deep-sea sensor networks, databases, and ship AIS data sources, and obtain deep-sea environment data and ship AIS data from various types of data sources through corresponding interfaces and protocols; Perform preprocessing operations such as data cleaning, calibration, and standardization on the obtained deep-sea environment data and ship AIS data; Match the preprocessed deep - sea environment data and vessel AIS data according to the timestamp and geographical location information, integrate data from different sources to form a deep - sea environment dataset containing multiple dimensions, and store the deep - sea environment dataset in the corresponding data warehouse.

3. A three-dimensional deep-sea environment deduction and simulation analysis system based on big data according to claim 2, characterized in that: The deep - sea three - dimensional scene modeling module specifically includes: Extract the deep - sea environment dataset from the data warehouse and classify the data in the deep - sea environment dataset, including seabed terrain data, water body data, biological data, and equipment data; For seabed terrain modeling, fuse terrain data with different resolutions, use terrain modeling tools to construct the seabed terrain, import the terrain data into modeling software, generate a terrain mesh through a height map, and then optimize the generated terrain mesh to adjust the level of detail of the terrain; For water body modeling, based on the physical properties of the water body, construct a physical model of the water body and use hydrodynamic simulation technology to simulate the dynamic effects of water flow, waves, and ocean currents; For biological modeling, according to the biological distribution data, construct a three - dimensional model library of marine organisms, including fish, plankton, and corals, use modeling software to create biological models, then place the biological models in the corresponding positions, simulate their behavior patterns, and combine the particle swarm optimization algorithm to simulate the group behavior and movement trajectories of organisms; For equipment modeling, construct a three - dimensional model library of deep - sea equipment, including research vessels, submersibles, and sensors, create three - dimensional models of deep - sea equipment according to the equipment positions and data, place the equipment models in the corresponding positions, and simulate their dynamic behaviors; Integrate the seabed terrain, water body, biological, and equipment elements into a unified three - dimensional scene, and optimize the entire scene, using LOD technology to dynamically adjust the level of detail of the models.

4. The three-dimensional deep-sea environment deduction and simulation analysis system based on big data according to claim 3, characterized in that: The rendering optimization module specifically includes: Configure Unreal Engine as the rendering engine for deep - sea three - dimensional environment deduction and simulation analysis, and initialize the rendering engine. Configure the basic parameters of the rendering engine, including rendering resolution, anti - aliasing settings, and shadow quality, and then load the deep - sea three - dimensional model and related resources into the rendering engine; Use the decimation tool of Unreal Engine to simplify the deep - sea three - dimensional model, reduce the number of faces and vertices, and in Unreal Engine, set multiple LOD levels for the model. According to the viewing distance and viewing angle of the model, dynamically adjust the level of detail of the model, and use the multi - level texture technology to automatically match the corresponding texture resolution according to the viewing distance; Configure the occlusion culling system of Unreal Engine to only render objects within the view frustum, ignore invisible objects, and enable geometry occlusion detection to cull the parts occluded by other objects, further reducing the rendering load; Use the PBR material system of Unreal Engine to optimize the reflection, refraction, and scattering parameters of the material, and combine the texture compression technology to achieve the streaming loading of textures and load texture data on demand; Enable screen - space global illumination and optimize the sampling of global illumination; Use hydrodynamic techniques to simulate the dynamic effects of water flow, waves, and ocean currents. At the same time, reduce the computational amount through a simplified model, use GPU acceleration technology to calculate the dynamic changes of water bodies in real time, and apply ray tracing technology to enhance the reflection and refraction effects of water bodies; Cache the animations of organisms and devices, and use skeletal animation technology to implement the dynamic behaviors of organisms and devices, optimizing skeletal weights and animation frame rates; Decompose the rendering tasks into multiple subtasks, utilize multi-threading technology to distribute the rendering tasks to multiple CPU cores for parallel processing, use GPU acceleration for rendering, and monitor the rendering performance in real time, including frame rate, CPU and GPU usage rates, and memory occupancy. Use performance analysis tools to locate performance bottlenecks.

5. A three-dimensional deep-sea environment deduction and simulation analysis system based on big data according to claim 1, characterized in that: The marine environment deduction and simulation module specifically includes: Use the three-dimensional model of the deep-sea environment generated by the rendering optimization module, including seabed terrain, water bodies, organisms, and device elements, and access real-time ship AIS data, including ship position, speed, and heading information; Combine the analysis results of the deep-sea environment analysis module including seabed terrain evolution trends, water body change trends, organism distribution and behavior trends, and device operation trends, and integrate historical deep-sea environment data with the analysis results; According to the three-dimensional model and real-time data at the current time point, initialize the state of the deep-sea environment, including the initial state of the seabed terrain, the initial conditions of the water body, the initial distribution and behavior patterns of organisms, and the initial positions and states of devices; Implement physical process simulations, including water body dynamics simulation, organism behavior simulation, and device dynamic simulation; Adopt a time-stepping method to gradually update the state of the deep-sea environment, and within each time step, update the state of the three-dimensional model according to the simulation results of the physical process; Based on the results of dynamic deduction, predict the change trends of the deep-sea environment in the future, realize long-term evolution prediction of the seabed terrain, seasonal changes and abnormal fluctuations prediction of water bodies, distribution and behavior change prediction of organisms, and operation state prediction of devices, and generate a scientific prediction report, including visual display and text description of the prediction results, and provide a scientific explanation of the prediction results.

6. The three-dimensional deep-sea environment deduction and simulation analysis system based on big data according to claim 5, wherein: The process of updating the state of the three-dimensional model is as follows: Define the time step, set it to 1 hour, and gradually advance the simulation time according to the set time step; Within each time step, update the state of the three-dimensional model according to the simulation results of the physical process, and update the seabed terrain, water body state, organism population distribution, and device state in the three-dimensional model according to the simulation results of the physical process and ecological process; Take the updated state as the initial condition for the next time step for iterative calculation, and after the simulation ends, output the required result data, including changes in the seabed terrain, seasonal changes and abnormal fluctuations of water bodies, distribution and behavior changes of organisms, and operation states of devices, and then perform visual processing on the result data.

7. A three - dimensional deep - sea environment deduction and simulation analysis system based on big data according to claim 6, characterized in that: The three-dimensional visualization display module specifically includes: Load the three-dimensional model of the deep-sea environment generated by the rendering optimization module, import the loaded three-dimensional model into the virtual scene, set the initial state of the scene, and configure the basic parameters of the rendering engine at the same time; Access real-time ship AIS data, including ship position, speed, and heading information, synchronize the AIS data to the ship and submersible models in the 3D model in real time, and use kinematic and dynamic models, combined with the equipment operation trend, to simulate the movement trajectories of the ship and submersible, and dynamically display the real-time movements of the ship and submersible in the 3D scene, including the navigation path and the diving trajectory; Provide a multi-scene resource switching function, enabling users to observe the deep-sea environment from different perspectives, including the global view, local view, and specific area view.

8. A three-dimensional deep-sea environment deduction and simulation analysis system based on big data according to claim 7, characterized in that: The deep-sea deduction typical scene roaming module specifically includes: According to user requirements, configure the relevant parameters of the roaming module, including the roaming speed, viewing angle range, and scene accuracy, and read the deep-sea scene data required for deduction from the database, including seabed topography, water body state, and biological distribution; Automatically navigate to the target location according to the location input by the user and start dynamic roaming. During the roaming process, dynamically update the rendering effect of the scene according to the user's viewing angle and position, provide real-time feedback on the user's operations, and at the same time provide a user interface that allows the user to independently adjust the roaming speed, supporting multiple speed modes including fast, medium, and slow; Dynamically display the deduction results in the 3D scene, highlight key information using different colors, icons, or animation effects, and provide a marking function that allows users to mark key locations in the 3D scene. The marking includes text annotations, icons, or color highlighting. At the same time, support users to quickly navigate to key locations through the marking, provide a query function, and users can input the location name or marking number to locate to the target location.

Images