A marine oil spill pollution diffusion early warning and prediction system and method

Abstract

The application discloses a kind of offshore oil spill pollution diffusion early warning prediction system and method, it is related to marine environment prediction technical field, the system includes: data acquisition and pretreatment unit, model construction and path simulation unit, path prediction and anomaly detection unit, model parameter optimization unit and visual display unit;Model construction and path simulation unit are used to construct offshore oil spill transport model based on marine numerical model, and the path of oil spill diffusion is simulated in combination with the multi-source data after pretreatment.The application widens the monitoring range of offshore oil spill event by collecting and pretreating multi-source data in marine environment, ensures the comprehensiveness and timeliness of data;At the same time, by constructing new characteristic variables, not only the accuracy of model simulation is enhanced, but also the prediction result is more in line with actual situation, so as to greatly improve the accuracy and reliability of prediction.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of marine environment prediction, in particular to a marine oil spill pollution diffusion early warning and prediction system and method. BACKGROUND

[0002] At present, the early warning and prediction of marine oil spill pollution mainly relies on satellite remote sensing monitoring, aircraft cruise reconnaissance, ship on-site patrol and buoy point monitoring and other means. Although these methods can provide information about oil spill events to some extent, they generally have problems such as limited monitoring range, low data update frequency and the need to improve monitoring accuracy. In addition, the existing mathematical model often fails to fully consider the complex interaction between wind field, wave field, tide and circulation and other environmental factors in simulating the process of oil spill drift and diffusion, which directly affects the accuracy and reliability of the prediction results.

[0003] In view of the problems in the related art, an effective solution has not been proposed yet. SUMMARY

[0004] In view of the problems in the related art, the present application proposes a marine oil spill pollution diffusion early warning and prediction system and method to overcome the above technical problems existing in the prior art.

[0005] To this end, the specific technical solutions adopted by the present application are as follows:

[0006] According to one aspect of the present application, a marine oil spill pollution diffusion early warning and prediction system is provided, which comprises a data acquisition and preprocessing unit, a model construction and path simulation unit, a path prediction and anomaly detection unit, a model parameter optimization unit and a visualization display unit.

[0007] The data acquisition and preprocessing unit is configured to acquire multi-source data in the marine environment and preprocess the acquired multi-source data.

[0008] The model construction and path simulation unit is configured to construct a marine oil spill transport model based on a marine numerical model, and simulate the path of oil spill diffusion in combination with the preprocessed multi-source data.

[0009] The path prediction and anomaly detection unit is configured to predict the oil spill diffusion path using a causal convolutional neural network model according to the simulation results of the oil spill diffusion path and historical multi-source data, and to detect abnormal values in the multi-source data in real time.

[0010] The model parameter optimization unit is configured to feed back the detected abnormal values to the causal convolutional neural network model, and to optimize the path prediction results of the oil spill diffusion by adjusting the parameters of the causal convolutional neural network model.

[0011] A visualization display unit is configured to visualize the optimized oil spill diffusion path prediction result by using a graph neural network technology.

[0012] In one embodiment, the data acquisition and preprocessing unit comprises a data acquisition module, a data cleaning module, and a feature engineering module.

[0013] The data acquisition module is configured to acquire multi-source data related to oil spill diffusion in the marine environment from a plurality of data sources.

[0014] The data cleaning module is configured to clean the acquired multi-source data to eliminate duplicate values and handle missing values.

[0015] The feature engineering module is configured to perform dimensionality reduction processing on the cleaned multi-source data by using a nonlinear dimensionality reduction algorithm, and to establish a multi-source data set based on the physical mechanism of oil spill diffusion.

[0016] In one embodiment, the multi-source data includes meteorological data, marine dynamic data, ship trajectories, historical oil spill case data, and oil physical and chemical properties.

[0017] In one embodiment, the feature engineering module comprises a parameter setting submodule, a data dimensionality reduction submodule, a feature calculation submodule, and a data set establishment submodule.

[0018] The parameter setting submodule is configured to set the parameters of the nonlinear dimensionality reduction algorithm based on the cleaned multi-source data.

[0019] The data dimensionality reduction submodule is configured to perform dimensionality reduction processing on the cleaned multi-source data by using the nonlinear dimensionality reduction algorithm, output the dimensionally reduced multi-source data, and establish an initial data set.

[0020] The feature calculation submodule is configured to analyze and calculate the kinetic potential of the oil spill point and the pollutant drift index based on the physical mechanism of oil spill diffusion.

[0021] The data set establishment submodule is configured to construct new feature variables based on the calculation results, add the new feature variables to the initial data set, and establish a multi-source data set.

[0022] In one embodiment, the model construction and path simulation unit comprises a parameter setting module, a model construction module, and a diffusion path simulation module.

[0023] The parameter setting module is configured to set the ocean region grid based on the ocean numerical model, discretize the control equation, and configure the boundary conditions to simulate the dynamic process of the ocean region.

[0024] The model construction module is configured to construct an offshore oil spill transport model in combination with the physical and chemical properties of the oil spill.

[0025] The diffusion path simulation module is used to simulate the motion trajectory of oil spill particles based on the constructed marine oil spill transport model and adopt the Lagrangian particle tracking algorithm, and generate the oil spill diffusion path by combining the pre-processed multi-source data.

[0026] In one embodiment, the model construction module includes: a characteristic analysis submodule, a horizontal transport model construction submodule, an initial model construction submodule, and a transport model construction submodule;

[0027] The characteristic analysis submodule is used to analyze the pre-processed multi-source data to obtain the physical and chemical characteristics of the oil spill during transportation;

[0028] The horizontal transport model construction submodule is used to establish an oil spill horizontal transport model based on the two-dimensional diffusion equation and the physical and chemical characteristics of the oil spill during the transport process;

[0029] The initial model construction submodule is used to combine the oil spill horizontal transport model with the three-dimensional diffusion equation to establish the initial offshore oil spill transport model;

[0030] The transport model construction submodule is used to perform spatial discretization processing on the initial offshore oil spill transport model using the finite difference method, and to construct the offshore oil spill transport model based on the spatial discretization processing results.

[0031] In one embodiment, the two-dimensional diffusion equation is formulated as:

[0032] ;

[0033] Where, ∂ represents partial differential; C Indicates oil film concentration; t Indicates time; λ Indicates the flow rate x Directional component; x Represents the spatial coordinate axis along the east-west direction; v Indicates the flow rate y Directional component; y Represents the spatial coordinate axis along the north-south direction; K represents the diffusion coefficient; ∇ 2 represents the Laplace operator.

[0034] In one embodiment, the three-dimensional diffusion equation is formulated as:

[0035] ;

[0036] Where, ∂ represents partial differential; C Indicates oil film concentration; t represents time; ∇ represents the divergence operator;γ represents the three-dimensional flow velocity vector; N represents the diffusion tensor; S Represents a source term.

[0037] In one embodiment, the path prediction and anomaly detection unit includes: a training set establishment module, a model construction and training module, a path prediction module, and an outlier detection module;

[0038] The training set establishment module is used to collect and pre-process historical multi-source data and establish a training set based on the simulation results of the oil spill diffusion path;

[0039] The model building and training module is used to build a causal convolutional neural network model based on the convolutional neural network algorithm and train the constructed causal convolutional neural network model using the training set;

[0040] The path prediction module is used to collect real-time multi-source data and input the real-time multi-source data into the trained causal convolutional neural network model to predict the oil spill diffusion path;

[0041] The outlier detection module is used to detect outliers on the input real-time multi-source data stream in combination with the isolation forest algorithm.

[0042] According to another aspect of the present invention, a method for early warning and prediction of marine oil spill pollution spread is provided, the method comprising:

[0043] S1. Collect multi-source data in the marine environment and pre-process the collected multi-source data;

[0044] S2. Based on the ocean numerical model, a marine oil spill transport model is constructed, and the oil spill diffusion path is simulated by combining pre-processed multi-source data;

[0045] S3. Based on the simulation results of the oil spill diffusion path and historical multi-source data, a causal convolutional neural network model is used to predict the oil spill diffusion path and detect outliers in the multi-source data in real time;

[0046] S4. Feeding the detected outliers back to the causal convolutional neural network model, and optimizing the oil spill diffusion path prediction results by adjusting the causal convolutional neural network model parameters;

[0047] S5. Use graph neural network technology to visualize the optimized oil spill diffusion path prediction results.

[0048] The beneficial effects of the present invention are:

[0049] 1、The present application widens the monitoring range of offshore oil spill events by collecting and preprocessing multi-source data in the marine environment, ensuring the comprehensiveness and timeliness of the data; at the same time, by constructing new characteristic variables, not only the accuracy of model simulation is enhanced, but also the prediction result is more in line with the actual situation, thereby greatly improving the accuracy and reliability of the prediction.

[0050] 2、The present application can quickly and accurately predict the oil spill diffusion path by using the causal convolutional neural network model, thereby fully utilizing historical data and simulation results to capture the dynamic characteristics of oil spill diffusion, and further realizing accurate prediction of the oil spill diffusion trend.

[0051] 3、The present application visualizes the optimized oil spill diffusion path prediction result by introducing the graph neural network technology, making the prediction result more intuitive and easy to understand, which helps decision makers quickly understand the trend and range of oil pollution diffusion, so as to take timely measures to reduce environmental pollution and economic losses. BRIEF DESCRIPTION OF DRAWINGS

[0052] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed in the embodiments will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor on the basis of these drawings.

[0053] Figure 1 is a principle block diagram of a marine oil spill pollution diffusion early warning and prediction system according to an embodiment of the present application;

[0054] Figure 2 is a flowchart of a marine oil spill pollution diffusion early warning and prediction method according to an embodiment of the present application.

[0055] In the drawings:

[0056] 1, data acquisition and preprocessing unit; 2, model construction and path simulation unit; 3, path prediction and anomaly detection unit; 4, model parameter optimization unit; 5, visual display unit. DETAILED DESCRIPTION

[0057] To further illustrate the embodiments, the present application provides drawings, which are part of the disclosure of the present application, mainly used to illustrate the embodiments, and can explain the operating principle of the embodiments in conjunction with the related description of the specification. With reference to these contents, those skilled in the art should understand other possible embodiments and advantages of the present application.

[0058] According to an embodiment of the present application, a marine oil spill pollution diffusion early warning and prediction system and method are provided.

[0059] The application will be further described in conjunction with the drawings and specific embodiments, as shown Figure 1 The offshore oil spill pollution diffusion early warning and prediction system according to the embodiment of the application comprises a data acquisition and preprocessing unit 1, a model construction and path simulation unit 2, a path prediction and anomaly detection unit 3, a model parameter optimization unit 4, and a visualization display unit 5.

[0060] The data acquisition and preprocessing unit 1 is configured to acquire multi-source data in a marine environment and preprocess the acquired multi-source data.

[0061] In this optional embodiment, the data acquisition and preprocessing unit 1 comprises a data acquisition module, a data cleaning module, and a feature engineering module.

[0062] The data acquisition module is configured to acquire multi-source data related to oil spill diffusion in a marine environment from a plurality of data sources.

[0063] In this optional embodiment, the multi-source data comprises meteorological data, marine dynamic data, ship trajectory data, historical oil spill case data, and oil spill physical and chemical properties.

[0064] It should be noted that meteorological data is one of the important parameters for predicting offshore oil spill diffusion, including wind speed, wind direction, air temperature, humidity, and other key information. These factors directly affect the drift and diffusion speed of oil spills in the ocean. For example, strong winds and specific wind directions can accelerate the diffusion of oil spills, while air temperature and humidity can affect the evaporation rate of oil spills.

[0065] Marine dynamic data mainly includes current speed, direction, water temperature, salinity, etc. These data are crucial for simulating and predicting the movement trajectory of oil spills in the ocean. For example, the direction and speed of the current directly affect the drift direction of the oil spill, while the water temperature can affect the solubility and biodegradation rate of the oil spill.

[0066] Ship trajectory data records the sailing path and speed of ships at sea, which is of great significance for determining the source of the oil spill event and the possible diffusion path. By analyzing the ship trajectory, the location and time of the oil spill event can be inferred, thereby more accurately predicting the diffusion trend of the oil spill.

[0067] Historical oil spill case data contains detailed information of past oil spill events, such as oil spill volume, oil spill type, diffusion speed, and impact range. These data provide valuable references and basis for predicting future oil spill events. By comparing and analyzing historical cases, the rules and trends of oil spill diffusion can be found, thereby developing more effective response strategies.

[0068] The physicochemical properties of spilled oil, including its density, viscosity, volatility, and solubility, are key factors in predicting its diffusion behavior. Different types of spilled oil have different physicochemical properties, which directly affect its behavior in the ocean. For example, oil with a lower density is more likely to float on the water surface, while oil with a higher viscosity may form a thicker oil film, affecting the diffusion speed.

[0069] The data cleaning module is used to clean the collected multi-source data to eliminate duplicate values and handle missing values.

[0070] The feature engineering module is used to perform dimensionality reduction processing on the cleaned multi-source data using a nonlinear dimensionality reduction algorithm, and to establish a multi-source data set based on the physical mechanism of oil spill diffusion.

[0071] In this optional embodiment, the feature engineering module includes a parameter setting submodule, a data dimensionality reduction submodule, a feature calculation submodule, and a data set establishment submodule.

[0072] The parameter setting submodule is used to set the relevant parameters of the nonlinear dimensionality reduction algorithm based on the cleaned multi-source data.

[0073] The data dimensionality reduction submodule is used to perform dimensionality reduction processing on the cleaned multi-source data using the nonlinear dimensionality reduction algorithm, output the dimensionally reduced multi-source data, and establish an initial data set.

[0074] The feature calculation submodule is used to analyze and calculate the kinetic potential of the oil spill point and the pollutant drift index based on the physical mechanism of oil spill diffusion.

[0075] It should be noted that the kinetic potential takes into account the effects of wind speed, water flow speed, wind direction, and other factors on oil spill diffusion.

[0076] The pollutant drift index is calculated based on water flow speed, wind direction, and oil spill area to determine the degree and direction of pollutant drift in water.

[0077] The data set establishment submodule is used to construct new feature variables based on the calculation results and add them to the initial data set to establish a multi-source data set.

[0078] It should be noted that in specific embodiments, the cleaned multi-source data contains 1000 samples, each with 20 features (including location, time, wind speed, etc.), as follows:

[0079] Step one, parameter setting

[0080] The t-SNE algorithm is selected, with a perplexity of 30, a learning rate of 200, and 1000 iterations.

[0081] Step two, dimensionality reduction processing

[0082] The 20-dimensional data is reduced to 2D using the t-SNE algorithm; the reduced data is used as the initial data set, containing 1000 samples, each with 2 features (reduced coordinates).

[0083] Step three, feature calculation

[0084] For each sample, calculate its kinetic potential = (wind speed 2) * 0.5.

[0085] Calculate the pollutant drift index = (water flow speed * wind direction cosine + wind speed * water flow direction cosine) * oil spill area.

[0086] Step four, multi-source data set establishment

[0087] Initial data set (reduced data): 1000 samples, each with 2 features.

[0088] New feature variables: kinetic potential (1000 values), pollutant drift index (1000 values).

[0089] Multi-source data set: 1000 samples, each with 4 features (2 features after dimensionality reduction + kinetic potential + pollutant drift index).

[0090] Model construction and path simulation unit 2, for constructing an offshore oil spill transport model based on a marine numerical model, and simulating the path of oil spill diffusion in combination with the preprocessed multi-source data.

[0091] In this optional embodiment, the model construction and path simulation unit 2 includes a parameter setting module, a model construction module, and a diffusion path simulation module.

[0092] The parameter setting module is used to set the ocean region grid based on the marine numerical model, discretize the control equation, and configure the boundary conditions to simulate the dynamic process of the ocean region.

[0093] The model construction module is used to construct an offshore oil spill transport model in combination with the physical and chemical properties of the oil spill.

[0094] In this optional embodiment, the model construction module includes a property analysis submodule, a horizontal transport model construction submodule, an initial model construction submodule, and a transport model construction submodule.

[0095] The characteristic analysis submodule is used to analyze the pre-processed multi-source data to obtain the physical and chemical characteristics of the oil spill during transportation.

[0096] The horizontal transport model construction submodule is used to establish an oil spill horizontal transport model based on the two-dimensional diffusion equation and combined with the physical and chemical characteristics of the oil spill during the transport process.

[0097] In this optional embodiment, the two-dimensional diffusion equation is formulated as:

[0098] ;

[0099] Where, ∂ represents partial differential; C Indicates oil film concentration; t Indicates time; λ Indicates the flow rate x Directional component; x Represents the spatial coordinate axis along the east-west direction; v Indicates the flow rate y Directional component; y Represents the spatial coordinate axis along the north-south direction; K represents the diffusion coefficient; ∇ 2 represents the Laplace operator.

[0100] The initial model construction submodule is used to combine the oil spill horizontal transport model with the three-dimensional diffusion equation to establish an initial offshore oil spill transport model.

[0101] In this optional embodiment, the three-dimensional diffusion equation is formulated as:

[0102] ;

[0103] Where, ∂ represents partial differential; C Indicates oil film concentration; t represents time; ∇ represents the divergence operator; γ represents the three-dimensional flow velocity vector; N represents the diffusion tensor; S Represents a source term.

[0104] The transport model construction submodule is used to perform spatial discretization processing on the initial offshore oil spill transport model using the finite difference method, and to construct the offshore oil spill transport model based on the spatial discretization processing results.

[0105] The diffusion path simulation module is used to simulate the motion trajectory of oil spill particles based on the constructed marine oil spill transport model and adopt the Lagrangian particle tracking algorithm, and generate the oil spill diffusion path by combining the pre-processed multi-source data.

[0106] It needs to be pointed out that based on the constructed offshore oil spill transport model, the Lagrangian particle tracking method algorithm is used to simulate the motion trajectory of the oil spill particles, and the preprocessed multi-source data is combined to generate the oil spill diffusion path, including:

[0107] Step one, parameter determination

[0108] Based on the constructed offshore oil spill transport model, the basic framework of the model is established, including the convection, diffusion, evaporation, dissolution, emulsification and other physical and chemical processes of the oil spill particles; the model parameters are set, such as wind speed, water flow speed, direction, oil film thickness, oil physical properties, etc.

[0109] Two, Lagrangian particle tracking method algorithm

[0110] (1) Disperse the oil film into a large number of oil particles, each oil particle represents a certain amount of oil; set the initial position, speed and concentration of each oil particle as attributes.

[0111] (2) Use the Lagrangian particle tracking method algorithm to simulate the motion trajectory of each oil particle according to the preprocessed multi-source data (such as wind speed, water flow speed, etc.); consider the convection transport and turbulent diffusion process of the oil particles, as well as the influence of the volatilization, dissolution, emulsification and other chemical processes of the oil particles on the motion trajectory.

[0112] (3) In each time step, update the position information of the oil particles according to their motion speed and direction; consider the interaction and collision between the oil particles, as well as the interaction between the oil particles and the marine environment (such as the oil particles adhering to the shore, etc.).

[0113] Three, generate oil spill diffusion path

[0114] (1) Record the motion trajectory of each oil particle, including position, time, speed, etc.; analyze the trajectory data to extract the key path and area of oil spill diffusion.

[0115] (2) Use GIS or visualization software to present the oil spill diffusion path in a graphical way. According to the needs, you can superimpose the marine environment data (such as sea current, wind speed, etc.), so as to more intuitively understand the relationship between oil spill diffusion and environmental factors.

[0116] The path prediction and anomaly detection unit 3 is used to predict the oil spill diffusion path by using the causal convolutional neural network model according to the simulation results of the oil spill diffusion path and the historical multi-source data, and to detect the abnormal values in the multi-source data in real time.

[0117] In this optional embodiment, the path prediction and anomaly detection unit 3 includes: a training set establishment module, a model construction and training module, a path prediction module and an abnormal value detection module;

[0118] The training set establishment module is configured to collect and pre-process historical multi-source data, combine path simulation results of oil spill diffusion, and establish a training set.

[0119] The model construction and training module is configured to construct a causal convolutional neural network model based on a convolutional neural network algorithm, and train the constructed causal convolutional neural network model using the training set.

[0120] The path prediction module is configured to collect real-time multi-source data, and input the real-time multi-source data into the trained causal convolutional neural network model to predict the diffusion path of the oil spill.

[0121] The anomaly value detection module is configured to detect anomaly values in the input real-time multi-source data stream using an isolation forest algorithm.

[0122] The model parameter optimization unit 4 is configured to feed back the detected anomaly values to the causal convolutional neural network model, and optimize the prediction results of the diffusion path of the oil spill by adjusting the parameters of the causal convolutional neural network model.

[0123] It should be noted that feeding back the detected anomaly values to the causal convolutional neural network model, and optimizing the prediction results of the diffusion path of the oil spill by adjusting the parameters of the causal convolutional neural network model includes:

[0124] Step 1, anomaly detection and feedback

[0125] (1) Mark and record the location and characteristics of the anomaly value, and analyze whether the anomaly value is caused by model prediction error, data noise or the particularity of the actual oil spill event; distinguish between real anomalies (such as changes in oil spill diffusion caused by sudden environmental events) and model errors.

[0126] (2) For anomaly values caused by model errors, adjust the model parameters; for real anomalies, consider including them in the training set to increase the generalization ability of the model.

[0127] Step 2, model parameter adjustment and optimization

[0128] According to the results of anomaly value analysis, adjust the hyperparameters of the model, such as the size and number of convolution kernels, learning rate, batch size, etc.; grid search, random search or Bayesian optimization can be used to find the optimal parameter combination.

[0129] The visualization display unit 5 is configured to use graph neural network technology to visualize the optimized prediction results of the diffusion path of the oil spill.

[0130] It should be noted that using graph neural network technology to visualize the optimized prediction results of the diffusion path of the oil spill includes:

[0131] Step one, construct a graph neural network model

[0132] (1) According to the physical mechanism and causal relationship of oil spill diffusion, design the structure of graph neural network; select appropriate graph neural network model, such as Graph Convolutional Networks (GCNs), Graph Attention Networks (GATs) and the like.

[0133] (2) Use the preprocessed data to train the graph neural network model.

[0134] Set appropriate loss function and optimizer, such as cross entropy loss function and Adam optimizer.

[0135] Through iterative training, optimize the model parameters and improve the prediction accuracy.

[0136] Step two, oil spill diffusion path prediction and optimization

[0137] (1) Use the trained graph neural network model to predict the oil spill diffusion path of real-time or historical data, and draw the oil spill diffusion path map according to the prediction results.

[0138] (2) Verify and evaluate the prediction results, such as calculating the prediction accuracy, recall rate and the like by using the verification set data; according to the evaluation results, adjust the model parameters or structure, and optimize the prediction performance.

[0139] Step three, visualization display

[0140] (1) Select appropriate visualization tools, such as TensorBoard, Matplotlib, Seaborn and the like, select appropriate chart types according to the visualization requirements, such as heat map, line chart, scatter plot and the like.

[0141] (2) According to the prediction results, draw the oil spill diffusion path map; mark the key positions and time points in the map, such as oil spill source, diffusion range, diffusion speed and the like, use different colors or marks to distinguish the diffusion situation of different time periods.

[0142] (3) Export the drawn oil spill diffusion path map to picture or video format; display the visualization results on the emergency response system or decision support platform, and provide intuitive information support for relevant personnel.

[0143] As shown in Figure 2 According to another embodiment of the present application, a method for predicting and warning the diffusion of offshore oil pollution is also provided, which comprises:

[0144] S1, collecting multi-source data in marine environment, and preprocessing the collected multi-source data;

[0145] S2, based on the marine numerical model, constructing an offshore oil spill transport model, and combining the pretreated multi-source data, simulating the path of oil spill diffusion;

[0146] S3, according to the simulation results of the oil spill diffusion path and the historical multi-source data, using a causal convolutional neural network model to predict the oil spill diffusion path, and detecting the abnormal values in the multi-source data in real time;

[0147] S4, feeding the detected abnormal values into the causal convolutional neural network model, adjusting the parameters of the causal convolutional neural network model to optimize the prediction results of the oil spill diffusion path;

[0148] S5, using graph neural network technology to visualize the optimized oil spill diffusion path prediction results.

[0149] In summary, with the help of the above technical solutions of the present application, by collecting and preprocessing multi-source data in the marine environment, the monitoring range of offshore oil spill events is widened, ensuring the comprehensiveness and timeliness of the data. At the same time, by constructing new feature variables, not only the accuracy of model simulation is enhanced, but also the prediction results are more in line with the actual situation, thereby greatly improving the accuracy and reliability of the prediction. By using the causal convolutional neural network model, the oil spill diffusion path can be quickly and accurately predicted, so as to fully utilize the historical data and simulation results to capture the dynamic characteristics of oil spill diffusion, and then realize the accurate prediction of the oil spill diffusion trend. By introducing the graph neural network technology to visualize the optimized oil spill diffusion path prediction results, the prediction results are more intuitive and easy to understand, which helps decision makers quickly understand the trend and range of oil pollution diffusion, so as to take timely measures to reduce environmental pollution and economic losses.

[0150] The above only describes the preferred embodiments of the present application, and does not limit the present application. Any modification, equivalent replacement, improvement, etc. within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A marine oil spill pollution spread warning and prediction system, characterized by: The system includes: data acquisition and preprocessing unit, model building and path simulation unit, path prediction and anomaly detection unit, model parameter optimization unit and visualization display unit; The data acquisition and preprocessing unit is used to acquire multi-source data in the marine environment and preprocess the acquired multi-source data; The model building and path simulation unit is used to build an offshore oil spill transport model based on the ocean numerical model, and simulate the oil spill diffusion path in combination with the pre-processed multi-source data; The path prediction and anomaly detection unit is used to predict the oil spill diffusion path based on the simulation results of the oil spill diffusion path and historical multi-source data using a causal convolutional neural network model, and to detect anomalies in the multi-source data in real time; The model parameter optimization unit is used to feed back the detected abnormal values ​​into the causal convolutional neural network model, and optimize the oil spill diffusion path prediction result by adjusting the causal convolutional neural network model parameters; The visualization display unit is used to visualize the optimized oil spill diffusion path prediction results using graph neural network technology; The data acquisition and preprocessing unit includes: a feature engineering module; The feature engineering module is used to perform dimensionality reduction processing on the multi-source data after data cleaning using a nonlinear dimensionality reduction algorithm, and to establish a multi-source data set based on the physical mechanism of oil spill diffusion; The feature engineering module includes: a parameter setting submodule, a data dimension reduction submodule, a feature calculation submodule and a data set establishment submodule; The parameter setting submodule is used to set relevant parameters of the nonlinear dimensionality reduction algorithm based on the multi-source data after data cleaning; The data dimensionality reduction submodule is used to perform dimensionality reduction processing on the multi-source data after data cleaning using a nonlinear dimensionality reduction algorithm, output the multi-source data after dimensionality reduction, and establish an initial data set; The feature calculation submodule is used to analyze and calculate the kinetic potential energy and pollutant drift index of the oil spill point based on the physical mechanism of oil spill diffusion; The data set establishment submodule is used to construct new feature variables based on the calculation results, and add the new feature variables to the initial data set to establish a multi-source data set; The path prediction and anomaly detection unit includes: a training set establishment module, a model construction and training module, a path prediction module and an outlier detection module; The training set establishment module is used to collect and pre-process historical multi-source data and establish a training set in combination with the oil spill diffusion path simulation results; The model construction and training module is used to construct a causal convolutional neural network model based on the convolutional neural network algorithm, and train the constructed causal convolutional neural network model using a training set; The path prediction module is used to collect real-time multi-source data and input the real-time multi-source data into the trained causal convolutional neural network model to predict the oil spill diffusion path; The outlier detection module is used to perform outlier detection on the input real-time multi-source data stream in combination with the isolation forest algorithm.

2. The marine oil spill pollution spread early warning and prediction system according to claim 1 is characterized in that: The data acquisition and preprocessing unit also includes: a data acquisition module and a data cleaning module; The data acquisition module is used to collect multi-source data related to oil spill diffusion in the marine environment from a plurality of data sources; The data cleaning module is used to clean the collected multi-source data to remove duplicate values ​​and process missing values.

3. The marine oil spill pollution diffusion early warning and prediction system according to claim 2 is characterized in that: The multi-source data includes: meteorological data, ocean dynamics data, ship trajectories, historical oil spill case data and physical and chemical properties of oil spills.

4. The marine oil spill pollution spread early warning and prediction system according to claim 1, characterized in that: The model building and path simulation unit includes: a parameter setting module, a model building module and a diffusion path simulation module; The parameter setting module is used to set the ocean area grid based on the ocean numerical model, discretize the control equations, and configure boundary conditions to simulate the dynamic process of the ocean area; The model building module is used to build an offshore oil spill transport model by combining the physical and chemical properties of the oil spill; The diffusion path simulation module is used to simulate the motion trajectory of oil spill particles based on the constructed offshore oil spill transport model using the Lagrangian particle tracking algorithm, and generate the oil spill diffusion path by combining pre-processed multi-source data.

5. The marine oil spill pollution spread warning and prediction system according to claim 4 is characterized in that: The model construction module includes: a characteristic analysis submodule, a horizontal transport model construction submodule, an initial model construction submodule and a transport model construction submodule; The characteristic analysis submodule is used to analyze the pre-processed multi-source data to obtain the physical and chemical characteristics of the oil spill during transportation; The horizontal transport model construction submodule is used to establish an oil spill horizontal transport model based on a two-dimensional diffusion equation and in combination with the physical and chemical characteristics of the oil spill during the transport process; The initial model construction submodule is used to combine the oil spill horizontal transport model with the three-dimensional diffusion equation to establish an initial offshore oil spill transport model; The transport model construction submodule is used to perform spatial discretization processing on the initial offshore oil spill transport model using the finite difference method, and construct the offshore oil spill transport model based on the spatial discretization processing result.

6. The marine oil spill pollution spread warning and prediction system according to claim 5, characterized in that: The formula of the two-dimensional diffusion equation is: ； Where, ∂ represents partial differential; C Indicates oil film concentration; t Indicates time; λ Indicates the flow rate x Directional component; x Represents the spatial coordinate axis along the east-west direction; v Indicates the flow rate y Directional component; y Represents the spatial coordinate axis along the north-south direction; K represents the diffusion coefficient; ∇ 2 represents the Laplace operator.

7. The marine oil spill pollution spread warning and prediction system according to claim 6, characterized in that: The formula of the three-dimensional diffusion equation is: ； Where, ∂ represents partial differential; C Indicates oil film concentration; t represents time; ∇ represents the divergence operator; γ represents the three-dimensional flow velocity vector; N represents the diffusion tensor; S Represents a source term.

8. A method for early warning and prediction of oil spill pollution spread at sea, for implementing the operation of the early warning and prediction system for oil spill pollution spread at sea according to any one of claims 1 to 7, characterized in that: The method includes: S1. Collect multi-source data in the marine environment and pre-process the collected multi-source data; S2. Based on the ocean numerical model, a marine oil spill transport model is constructed, and the oil spill diffusion path is simulated by combining pre-processed multi-source data; S3. Based on the simulation results of the oil spill diffusion path and historical multi-source data, a causal convolutional neural network model is used to predict the oil spill diffusion path and detect outliers in the multi-source data in real time; S4. Feeding the detected outliers back to the causal convolutional neural network model, and optimizing the oil spill diffusion path prediction results by adjusting the causal convolutional neural network model parameters; S5. Use graph neural network technology to visualize the optimized oil spill diffusion path prediction results.

Images