A pollutant dispersion simulation and assessment method for marine environmental protection

Abstract

The present invention discloses a pollutant diffusion simulation and assessment method for marine environmental protection, which belongs to the field of marine pollutant technology and includes the following steps: configuring a pollutant model based on obtained multi-source data, initializing dynamic factors, obtaining initialization data such as ocean current speed and direction, tidal height and period, wind field intensity and direction, temperature, salinity, and pH value, and defining a chemical reaction model based on the initialization data combined with the pollutant characteristic matrix. Through the fusion and dynamic modeling of multi-source data, the simulation accuracy of the pollutant diffusion path is significantly improved, and the dynamic change characteristics of the diffusion process can be accurately captured. Combined with the efficient initialization of dynamic factors and the multi-dimensional analysis of physical and chemical conditions, a comprehensive analysis of pollutant diffusion behavior is achieved, providing reliable technical support for marine pollution monitoring and assessment.

Description

Technical Field

[0001] The present invention belongs to the technical field of marine pollutants and specifically relates to a pollutant diffusion simulation and assessment method for marine environmental protection. Background Art

[0002] The ocean is one of the most important ecosystems on Earth. It not only provides humans with rich resources and a living environment, but also plays a key role in global climate regulation and ecological balance. However, with the acceleration of global industrialization and urbanization, marine pollution problems are becoming increasingly serious. For example, industrial wastewater, agricultural runoff, oil spills, and large amounts of plastic waste enter the ocean, posing a serious threat to the marine ecosystem. The spread and migration of pollutants not only destroy the living environment of marine organisms, but also pose a major hidden danger to human economic activities and health, becoming one of the core challenges of global environmental governance.

[0003] The invention patent with announcement number [CN118657091A] discloses a method and system for assessing the degree of marine pollution based on a pollutant diffusion model. Delft 3D software is used to establish a three-dimensional hydrodynamic model to simulate the hydrodynamic characteristics of the target sea area. By verifying and analyzing the spatiotemporal changes in the target sea area's tide level, current, temperature, and salinity, the water quality model parameters suitable for the target area are determined. A three-dimensional water quality model is constructed based on the hydrodynamic model to obtain the main water quality parameters of the target sea area for constructing an assessment matrix. The water quality quantitative assessment value obtained by the superior and inferior solution distance method is used. The time series data of the water quality quantitative assessment value of each monitoring point within the monitoring period λ is used as a sample for classification processing to obtain the pollution degree level at the monitoring point in the target sea area.

[0004] However, as a multi-dimensional, dynamic and complex process, pollutant diffusion is affected by a variety of dynamic factors such as ocean currents, tides, and wind fields, as well as physical and chemical conditions such as temperature, salinity and pH. Its diffusion path and laws are difficult to describe with a simple model. Existing technologies are insufficient in extracting the characteristics of short-term diffusion and long-term diffusion, and the combined analysis of high-frequency and low-frequency features has limitations. It is difficult to accurately identify key high-risk areas, which limits the effectiveness of pollution control. Summary of the Invention

[0005] In response to the shortcomings of existing solutions in the prior art, such as insufficient feature extraction of short-term diffusion and long-term diffusion, and limitations in the combined analysis of high-frequency and low-frequency features, this paper proposes a pollutant diffusion simulation and assessment method for marine environmental protection. The method aims to accurately capture the multi-scale diffusion characteristics of pollutants and scientifically identify high-risk areas by fusing multi-source data, combining short-term and long-term diffusion features, multi-scale feature extraction and frequency domain analysis technology.

[0006] The present invention is implemented by adopting the following technical solution: a pollutant diffusion simulation and assessment method for marine environmental protection, comprising the following:

[0007] Step S1: Acquire multi-source data and perform denoising, outlier detection and correction, and data interpolation and reconstruction preprocessing operations on them in sequence to generate a database;

[0008] The multi-source data includes dynamic factor data, physical and chemical condition data and pollutant characteristic data, and the pre-processed multi-source data generates a dynamic factor data matrix, a physical and chemical condition matrix and a pollutant characteristic matrix;

[0009] Step S2, constructing a pollutant concentration field: configuring a pollutant model based on the processed multi-source data, initializing the dynamic factors to obtain initialization data, and defining a chemical reaction model in combination with the pollutant characteristic matrix; determining the pollutant concentration field based on the pollutant model and the chemical reaction model; the initialization data includes ocean current speed and direction, tidal height and period, wind field intensity and direction, temperature, salinity, and pH value;

[0010] Step S3, generating a multi-scale feature description: decomposing the pollutant concentration field into short-term diffusion and long-term diffusion, and obtaining the corresponding short-term diffusion characteristics and long-term diffusion characteristics, then extracting the microscopic and macroscopic characteristics of the pollutant concentration field, and combining the short-term diffusion characteristics and long-term diffusion characteristics to generate a multi-scale feature description;

[0011] Step S4, optimizing the pollutant model based on multi-scale feature description;

[0012] Based on the aforementioned description of pollutant concentration fields and multi-scale characteristics, the pollutant model is dynamically adjusted in combination with the monitoring concentration field data to optimize the calculation accuracy of the pollutant concentration field. The rapid transport path of pollutants is extracted based on the short-term diffusion characteristics, and the pollutant concentration field is dynamically adjusted accordingly. The diffusion rate of pollutants in different grid cells is calculated based on the long-term diffusion characteristics, and the diffusion coefficient is corrected accordingly. The reaction rate constant is adjusted in combination with the physical and chemical condition data. The sensor network is used to monitor the distribution of pollutants to obtain a real-time monitoring concentration field, which is compared with the pollutant concentration field calculated by the pollutant model to calculate the error. Based on the error analysis, if the error exceeds the set threshold, the dynamic field, diffusion parameters and reaction rate constant are adjusted, and iterative optimization is triggered.

[0013] Step S5, draw a pollutant diffusion path map and generate a risk assessment report: simulate the diffusion path of pollutants through the optimized pollutant model to obtain the pollutant path, define high-risk areas based on the obtained pollutant diffusion path map, compare the obtained monitored pollutant concentration field with the pollutant concentration field obtained by the optimized pollutant model, draw a pollutant diffusion path map, highlight key areas on the diffusion path map, and generate a risk assessment report.

[0014] Step S6: Set control measures based on the risk assessment report and the pollutant diffusion path map, evaluate the control effect in combination with the control effect feedback, adjust the control measures, and finally generate a decision support report.

[0015] Compared with the prior art, the advantages and positive effects of the present invention are:

[0016] (1) Through the fusion and dynamic modeling of multi-source data, the simulation accuracy of pollutant diffusion paths has been significantly improved, and the dynamic characteristics of the diffusion process have been accurately captured. Combined with the efficient initialization of dynamic factors and the multidimensional analysis of physical and chemical conditions, a comprehensive analysis of pollutant diffusion behavior has been achieved, providing reliable technical support for marine pollution monitoring and assessment.

[0017] (2) Based on the integration of data on multidimensional dynamic factors such as ocean currents, tides, and wind fields, as well as physical and chemical conditions such as temperature, salinity, and pH, the model's adaptability to the actual ocean environment is significantly improved through denoising, interpolation, and dynamic correction techniques. The precise processing of data fusion ensures the efficiency of dynamic factor initialization, laying the foundation for the reliability of simulation results.

[0018] (3) By combining short-term and long-term diffusion characteristics, as well as frequency domain analysis and spatial decomposition techniques, the macroscopic and microscopic characteristics of the pollutant concentration field are fully extracted. The dynamic analysis method can simultaneously analyze high-frequency and low-frequency characteristics, solving the problem of inaccurate identification of high-risk areas in complex sea areas.

[0019] (4) Through multi-scale feature extraction and dynamic adjustment models, pollutant diffusion paths and risk areas are accurately identified. Combined with the generated diffusion path map and risk assessment report, it provides a scientific basis for pollution prevention and control and optimal resource allocation, and helps promote the refined management of marine pollution control and the sustainable development of the ecosystem. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 This is a flow chart of the pollutant diffusion simulation and assessment method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0021] In order to more clearly understand the above-mentioned objects, features and advantages of the present invention, the present invention is further described below with reference to the accompanying drawings and embodiments. In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention can also be implemented in other ways than those described herein. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0022] Embodiment, this embodiment provides a pollutant diffusion simulation and assessment method for marine environmental protection, such as Figure 1 Said method comprises the following steps:

[0023] Step S1: Acquire multi-source data and perform preprocessing operations on it to generate a database;

[0024] The multi-source data includes dynamic factor data, physical and chemical condition data and pollutant characteristic data, and the pre-processed multi-source data generates a dynamic factor data matrix, a physical and chemical condition matrix and a pollutant characteristic matrix;

[0025] Step S2: performing calculations based on the pollutant model and the chemical reaction model to obtain a pollutant concentration field;

[0026] The pollutant model is configured based on the preprocessed multi-source data, and the dynamic factor is initialized. The data after dynamic factor initialization is combined with the pollutant characteristic matrix to construct a chemical reaction model, thereby obtaining the pollutant concentration field;

[0027] Step S3, generating multi-scale feature description;

[0028] The pollutant concentration field is decomposed into short-term diffusion and long-term diffusion, and the corresponding short-term diffusion characteristics and long-term diffusion characteristics are obtained. Then, the microscopic and macroscopic characteristics of the pollutant concentration field are extracted, and the short-term diffusion characteristics and long-term diffusion characteristics are combined to generate a multi-scale feature description;

[0029] Step S4, optimizing the pollutant model based on multi-scale feature description;

[0030] Monitor pollutant distribution through sensor networks and obtain pollutant concentration fields. By comparing monitoring data with model calculation results, identify errors in the pollutant diffusion model and dynamically optimize the pollutant model.

[0031] Step S5: Simulate the diffusion path of pollutants using a pollutant model, compare the pollutant concentration field obtained by monitoring with the pollutant path, draw a pollutant diffusion path map, highlight high-risk areas on the diffusion path map, and generate a risk assessment report;

[0032] Step S6: Set control measures based on the risk assessment report and the pollutant diffusion path map, evaluate the control effect in combination with the control effect feedback, adjust the control measures, and finally generate a decision support report.

[0033] The following is a detailed description of this solution in conjunction with specific implementation methods:

[0034] In step S1, the multi-source data obtained include dynamic factor data, physical and chemical condition data, and pollutant characteristic data; wherein the dynamic factor data includes the ocean current speed and direction recorded by the current meter, the tidal height and period continuously recorded by the water level sensor, and the wind field intensity and direction obtained by the wind speed and direction meter. The physical and chemical condition data include the temperature of the sea surface and different depths measured by the water temperature sensor, the seawater conductivity detected by the salinity sensor, and the pH value at different points measured by the pH sensor. The pollutant characteristic data includes the graded measurement of the pollutant concentration in the pollution source area using a water quality analyzer, marked as the initial input concentration, the detection of the actual diffusion boundary of the pollutant through multi-point sampling, the collection of water samples and the analysis of the dissolution and adsorption behavior of the pollutants under different temperature and salinity conditions in the laboratory to generate a database.

[0035] When preprocessing multi-source data, the main operations include denoising, outlier detection and correction, and data interpolation and reconstruction. Denoising uses the Kalman filter algorithm to denoise the collected ocean current speed and direction data to eliminate the environmental noise interference on the sensor. Outlier detection and correction uses the box plot method to detect outliers, mark abnormal tidal height and wind speed data as abnormal data points, and correct unreasonable data detected in the temperature and salinity distribution through interpolation correction. Data interpolation and reconstruction uses the Lagrange interpolation method to fill in the gaps in the collected data, especially in areas not covered by the monitoring network, to reconstruct the complete data set, and generate the dynamic factor data matrix, physical and chemical condition matrix and pollutant characteristic matrix for the corresponding multi-source data. It should be noted here that since preprocessing is a technology well known to people in this field, it will not be described in detail here.

[0036] In the specific implementation process of step S2, a pollutant model is configured based on the obtained multi-source data, and a dynamic factor is initialized based on the configured pollutant model to obtain ocean current speed and direction, tidal height and period, wind field intensity and direction, temperature, salinity and pH value. Based on the obtained ocean current speed and direction, tidal height and period, wind field intensity and direction, temperature, salinity and pH value combined with the pollutant characteristic matrix, a chemical reaction model is defined. Based on the configured pollutant model and the chemical reaction model, calculations are performed to obtain the pollutant concentration field, specifically:

[0037] Step S21: configure a pollutant model based on the obtained multi-source data, using the multidimensional dynamics of pollutants as the mathematical basis of the pollutant model, specifically:

[0038]

[0039] Where: C is the pollutant concentration, which represents the concentration distribution at any time and space position during the diffusion process; t is the time, which describes the dynamic evolution of pollutant diffusion. is the convection term, which represents the transport effect of the dynamic factor on the pollutant concentration. is the gradient operator, which represents the spatial change rate, and v is the dynamic field vector, which represents the flow speed of pollutants under the influence of ocean currents, wind fields, etc. is the diffusion term, which represents the molecular diffusion and turbulent diffusion of pollutants in water bodies. D is the diffusion coefficient, which reflects the diffusion rate and depends on the solubility of pollutants and environmental conditions. is the Laplace operator, which represents the second-order rate of change of concentration in space. R(C,T,S,pH) is the chemical reaction term, which describes the transformation behavior of pollutants under physical and chemical conditions. T is temperature, S is salinity, and pH is acidity.

[0040] The pollutant model divides the multidimensional space and decomposes the area into multiple grid cells. Each cell contains specific dynamic factors and physical and chemical conditions, with parameters of 500 meters × 500 meters × 10 meters per grid.

[0041] Step S22: Based on the grid structure of the pollutant model, the dynamic factors are assigned and calculated to establish a spatiotemporal continuous dynamic field. Under the action of the dynamic field, the pollutant concentration field is calculated in combination with the chemical reaction model;

[0042] (1) Extract ocean current speed and direction, tidal height and period, and wind field strength and direction from the dynamic factor data matrix. Ocean current speed and direction are the dominant factors in the dynamic field, configuring the velocity vector distribution within the region. Tidal height and period are used to describe the cyclical changes in tide-induced velocity. Wind field strength and direction are used to influence the drift characteristics of surface pollutants.

[0043] In each grid cell, the vector field is calculated based on the ocean current speed and direction, tidal height and period, and wind field intensity and direction to form a spatiotemporal continuous dynamic field. Model verification is then performed to verify the physical laws of the dynamic field region, including the continuity of dynamic field changes between adjacent grid cells. The dynamic field calculation results are then compared with historical monitoring data for error analysis, and the mean square error is used for comparison.

[0044] Extract temperature, salinity, and pH from the physicochemical condition matrix. Temperature is used to influence diffusion coefficients and reaction rates, salinity is used to influence solubility and migration behavior, and pH is used to influence adsorption behavior and chemical reaction pathways.

[0045] (2) Based on the physicochemical environmental parameters and pollutant characteristic matrix, a chemical reaction model is defined, specifically:

[0046] R(C,T,S,pH)=k(T,S,pH)·C

[0047] Where: R(C,T,S,pH) is the chemical reaction term, which represents the chemical transformation rate of pollutants such as decomposition and adsorption under different physicochemical conditions. k(T,S,pH) is the chemical reaction rate constant, which depends on temperature (T), salinity (S), and pH. Temperature (T) affects the reaction rate, salinity (S) affects the solubility of pollutants, and pH determines the adsorption characteristics and reaction path. C is the pollutant concentration, which is the key input variable for the chemical reaction.

[0048] The initial pollutant concentration is obtained from the pollutant characteristic matrix, and the initial pollutant distribution is set in the gridded area. Based on the monitoring data of the pollution source area, the initial pollutant concentration is determined and mapped to the pollutant characteristic matrix, and the concentration gradient of the initial pollutant is set. In the gridded area, the initial pollutant concentration is assigned to each calculation unit, and the initial reaction rate of each grid unit is set in combination with the chemical reaction model of the pollutant, including: open boundary, fixed boundary and reflection boundary. The open boundary is: allowing pollutants to diffuse with the dynamic factor without imposing concentration constraints. The fixed boundary is: setting the pollutant concentration constant to simulate the continuous input or sedimentation process of pollutants. The reflection boundary is: preventing pollutants from escaping the calculation domain, reflecting the restrictive effect of the actual ocean topography. Finally, a complete initial pollutant concentration field is generated.

[0049] According to the pollutant characteristic matrix, the effective range of the initial pollutant concentration is obtained, and data points in irrelevant areas are eliminated. Based on the pollutant concentration threshold, the effective pollutant diffusion area is set, and areas with low background pollutant concentration or unpolluted are eliminated. Combined with the geographical information of the pollution source, only the grid cells affected by the pollution are retained, and the boundary diffusion characteristics of the pollutants are defined, including: open boundary, fixed boundary and reflection boundary. The open boundary means that pollutants can freely diffuse out of the simulation area with the ocean current without imposing concentration constraints. The fixed boundary means that the pollutant concentration is set to a fixed value to simulate the continuous input or deposition of pollutants. The reflection boundary means that pollutants are not allowed to diffuse out of the area, simulating the impact of coastlines or breakwaters. The chemical reaction rate is adjusted based on physical and chemical conditions. In the boundary area, the chemical reaction rate of pollutants is dynamically adjusted according to the spatial distribution of temperature, salinity and pH value. In areas with higher temperatures, the degradation rate of pollutants is accelerated. In areas with higher salinity, the solubility and adsorption behavior of pollutants are adjusted. Under different pH conditions, the redox reaction rate of pollutants is corrected.

[0050] (4) Perform calculations using the configured pollutant model and chemical reaction model;

[0051] The grid structure of the pollutant model is used to establish a spatiotemporal continuous dynamic field, allowing pollutants to be transported and diffused under the influence of ocean currents, tides, and wind field dynamic factors, thereby obtaining a pollutant concentration field, specifically:

[0052]

[0053] Where: C(x,y,z,t) is the pollutant concentration field, which represents the pollutant concentration distribution at any time t and spatial position (x,y,z). It is the rate of change of the concentration field relative to time, indicating the growth or decay rate of the pollutant concentration over time. is the convection term, which represents the transport effect of the dynamic factor on the pollutant concentration. is the gradient operator, describing the spatial rate of change of concentration, v is the dynamic field vector, (vC) is the concentration distribution under the dynamic field, is the diffusion term, which represents the molecular diffusion and turbulent diffusion of pollutants in water bodies. D is the diffusion coefficient, which is determined by the solubility of pollutants and physical and chemical conditions such as temperature and salinity. is the Laplace operator, describing the spatial second-order rate of change of concentration, R(C(m), T(m), S(m), pH) is the chemical reaction term, describing the transformation behavior of pollutants under physical and chemical conditions, T (m) It is temperature, which affects the reaction rate, S(m) is salinity, which affects the solubility of pollutants, pH is acidity and alkalinity, which determines the adsorption characteristics and reaction path, and C(m) is concentration, which is the core variable of chemical reactions.

[0054] In the specific implementation process of step S3, the obtained pollutant concentration field is decomposed into short-term diffusion and long-term diffusion to obtain short-term diffusion characteristics and long-term diffusion characteristics. The time signal of the pollutant concentration field is subjected to frequency domain analysis using fast Fourier transform to extract high-frequency components and obtain microscopic characteristics. The pollutant concentration field is decomposed into characteristics of multiple spatial resolutions using wavelet transform to obtain macroscopic characteristics. The short-term diffusion characteristics, long-term diffusion characteristics, microscopic characteristics and macroscopic characteristics are combined to generate a multi-scale feature description, which specifically includes:

[0055] Step S31: Decompose the obtained pollutant concentration field into short-term diffusion and long-term diffusion, specifically:

[0056] C(x,y,z,t)=C 短期 (x,y,z,t)+C 长期 (x,y,z,t)

[0057] Where: C(x,y,z,t) is the pollutant concentration field, which represents the pollutant concentration distribution at any time t and spatial position (x,y,z), C 短期 (x, y, z, t) is the short-term diffusion characteristic, which describes the instantaneous fluctuation of pollutants in a short period of time, such as the periodic concentration fluctuation caused by tidal changes, C 长期 (x, y, z, t) is the long-term diffusion characteristic, which represents the average concentration distribution trend of the pollutant over a longer time scale.

[0058] Step S32: Use fast Fourier transform to perform frequency domain analysis on the time signal of the pollutant concentration field, extract the high-frequency components, calculate the long-term change trend of the pollutant concentration field by time-weighted average method, and obtain microscopic characteristics. The microscopic characteristics are used to capture the rapid changes in the concentration of pollutants in local areas. Specifically,

[0059]

[0060] Where: C 长期 (x, y, z) is the long-term diffusion characteristic, which represents the average concentration distribution within the time range T, where T is the total time range of the long-term observation. C(x, y, z, t) is the pollutant concentration field, which represents the pollutant concentration that changes with time. It is a time integration operation, which is used to calculate the cumulative value of the concentration from time 0 to time T. is the time averaging factor used to calculate the average value within the observation range.

[0061] Wavelet transform is used to decompose the pollutant concentration field into multiple spatial resolution features to obtain macroscopic features. Macroscopic features are used to reflect the overall diffusion range of pollutants, specifically:

[0062]

[0063] Where: C(x,y,z) is the pollutant concentration field, which represents the pollutant concentration distribution at any spatial point (x,y,z), C j (x, y, z) is the characteristic component at the jth spatial resolution scale, j = 0 is the low-frequency component, reflecting the macroscopic global concentration distribution, j = N is the high-frequency component, indicating the rapid change of local concentration, and N is the total number of decomposition scales, which is usually selected according to the size of the study area and the required resolution. It is a superposition operation, which means reconstructing the characteristic components of different scales into a complete concentration field.

[0064] Step S33: Combine the short-term diffusion features, long-term diffusion features, microscopic features, and macroscopic features to generate a multi-scale feature description, specifically:

[0065]

[0066] Where: C 多尺度 (x, y, z, t) is a multi-scale feature description that combines the concentration variation characteristics of time and space. j (x, y, z) is the characteristic component of the jth spatial scale, describing the concentration distribution of pollutants at different resolutions, C 短期 (x, y, z, t) is the short-term diffusion characteristic, which indicates the rapid fluctuation of pollutants in a short period of time. 长期 (x,y,z) is the long-term diffusion characteristic, describing the average concentration distribution of pollutants, Represents the superposition of all spatial scales.

[0067] This embodiment is based on multi-scale feature description and assigns weights to pollutant diffusion characteristics at different time scales and spatial scales. In the initial stage of pollutant diffusion, short-term diffusion characteristics and microscopic characteristics have higher priority and are mainly used to characterize local changes of pollutants in a short period of time. In the long-term evolution stage of pollutants, long-term diffusion characteristics and macroscopic characteristics have higher weights and are used to describe the diffusion trend of pollutants in a large area and guide the overall pollutant diffusion prediction. Through the multi-scale feature fusion method, short-term diffusion characteristics and microscopic characteristics are superimposed with long-term diffusion characteristics and macroscopic characteristics to characterize short-term local changes. Combined with the time series data of the pollutant concentration field, interpolation methods and filtering algorithms are used to smooth short-term high-frequency fluctuations and maintain the continuity of long-term trends. Visualization technology is used to generate dynamic diffusion maps of pollutants at different time scales and spatial scales, showing the diffusion path, concentration changes and high-risk areas of pollutants. Short-term diffusion characteristics are mainly used for high-frequency fluctuations in the early stage of pollutant leakage, and to predict the transport direction and diffusion speed of pollutants in a short period of time. Long-term diffusion characteristics are used to analyze the global diffusion trend of pollutants, determine the impact range of pollutants in the long term in the future, and evaluate the cumulative effect of pollutants. Ultimately, the generated dynamic diffusion map comprehensively displays the short-term changes, high-frequency fluctuation areas and long-term diffusion trends of pollutants.

[0068] In the specific implementation process of step S4, the pollutant distribution is monitored through the sensor network, and the monitoring pollutant concentration field is obtained. By comparing the monitoring data with the model calculation results, the error of the pollutant diffusion model is identified, and the pollutant model is dynamically optimized. Specifically,

[0069] Step S41: extracting the rapid transport path of pollutants based on the obtained short-term diffusion characteristics, and dynamically adjusting the pollutant concentration field, specifically:

[0070] v 优化 =v 初始 +Δv

[0071] Where: vinitial is the initial dynamic field vector, generated by the initialized ocean current, tide and wind field data, Δv is the correction value, which is obtained by monitoring data and short-term diffusion characteristics C 短期 (x,y,z,t) calculations are used to adjust the flow rate and direction of the pollutant concentration field.

[0072] The diffusion rate of pollutants in different grid cells is calculated based on the long-term diffusion characteristics, and the diffusion coefficient is corrected. Specifically,

[0073]

[0074] Where: D 优化 is the optimized diffusion coefficient, which represents the average rate of pollutants in each grid cell, D 初始 is the initial diffusion coefficient, set by the physicochemical properties of the pollutant, Is the long-term diffusion characteristic C 长期 The time rate of change of (x,y,z) indicates the rate of change of pollutant concentration in the long-term diffusion trend.

[0075] The reaction rate constant is adjusted based on the physicochemical conditions data, specifically:

[0076] R 优化 (C(m),T(m),S(m),pH)=k 优化 (T(m),S(m),pH)·C(m)

[0077] Where: R 优化 (C(m), T(m), S(m), pH) is the optimized chemical reaction term, which represents the chemical transformation rate of pollutants in each grid unit. k optimization (T(m), S(m), pH) is the optimized chemical reaction rate constant, which is affected by physical and chemical conditions. T(m) is temperature, which affects the chemical reaction rate. S(m) is salinity, which affects the solubility of pollutants. pH is acidity and alkalinity, which determines the adsorption characteristics and reaction path of pollutants. C(m) is the pollutant concentration, which represents the initial concentration of pollutants in each grid unit.

[0078] Step S42: Monitor the pollutant distribution through the sensor network and obtain the monitored concentration field. Compare the monitored concentration field with the pollutant concentration field obtained by the pollutant model calculation to calculate the error. Specifically,

[0079] ∈(x,y,z,t)=C 监测 (x,y,z,t)-C 模拟 (x,y,z,t)

[0080] Where: ∈(x,y,z,t) is the error distribution, which represents the difference between the monitoring concentration field and the pollutant concentration field, C 监测(x, y, z, t) is the concentration field monitored in real time, which comes from sensors and remote sensing data. 模拟 (x,y,z,t) is the concentration field calculated by the model, which is generated by the optimized dynamic field, diffusion coefficient, and chemical reaction terms.

[0081] By comparing the pollutant concentration field with the one calculated by the pollutant model, an iterative optimization process is triggered if the error exceeds a dynamically set threshold. The error threshold is adaptively adjusted based on pollutant type, ocean dynamics, and monitoring equipment accuracy, with a typical setting range of 2%-5%. The error threshold is optimized by statistically analyzing the errors between historically monitored pollutant concentration field data and the pollutant concentration field calculated by the pollutant model. Iterative optimization incorporates the latest monitored ocean current velocity, tidal height, and wind intensity data to locally correct the dynamic field underlying pollutant transport. The velocity vector field is optimized using the least squares method. In areas with large errors, the diffusion coefficient in the pollutant diffusion model is dynamically adjusted to match the actual monitored pollutant concentration field data, and interpolation is used to smooth the diffusion coefficient between regions. Based on the latest temperature, salinity, and pH data, the chemical reaction rate constants in the pollutant model are optimized. Pollutant degradation rates are adjusted in high-temperature regions, and pollutant dissolution and adsorption equilibrium parameters are optimized in high-salinity regions. Pollutant redox reaction rates are corrected under different pH conditions to ensure that the pollutant diffusion calculations more closely reflect the actual chemical transformation behavior in the marine environment. Finally, the pollutant concentration field obtained by the pollutant model calculation is recalculated based on the optimized dynamic field on which the pollutant concentration field depends, the diffusion coefficient of the pollutant model calculation, and the conversion rate of the chemical reaction term.

[0082] In the specific implementation process of step S5, the pollutant diffusion path is simulated through the pollutant model to obtain the pollutant path. Based on the pollutant diffusion path and pollutant concentration field, the pollutant diffusion risk is assessed, high-risk areas are identified, and a pollutant diffusion path map is drawn to generate a risk assessment report, which specifically includes:

[0083] Step S51: Use the pollutant model to simulate the diffusion path of the pollutants and obtain the pollutant path, specifically:

[0084]

[0085] Where: C(x,y,z,t) is the pollutant concentration field, which represents the pollutant concentration distribution at any time t and spatial position (x,y,z). is the rate of change of concentration over time t, describing the dynamic process of increase or decrease of pollutant concentration. is the convection term, which represents the transport effect of the optimized dynamic field on the pollutant concentration. is the gradient operator, which represents the spatial rate of change, v优化 is the optimized dynamic field vector, describing the effects of ocean currents, tides, and wind fields on pollutant transport, (v optimized C) is the concentration transport under the action of the dynamic field, is the diffusion term, which represents the optimized diffusion coefficient D 优化 , the diffusion effect of pollutants in space, is the Laplace operator, which represents the second-order rate of change of concentration in space, R 优化 (C(m), T(m), S(m), pH) is a chemical reaction term that describes the decomposition, adsorption, and other behaviors of pollutants under optimized physicochemical conditions.

[0086] Based on the pollutant path, the diffusion path of pollutants is extracted, the points where the concentration value reaches the warning threshold are marked, and their diffusion direction and speed are tracked to generate a pollutant diffusion path map, showing the diffusion trajectory of pollutants from the source to the surrounding areas.

[0087] Step S52: defining high-risk areas based on the obtained pollutant diffusion path map;

[0088] When the pollutant concentration in the pollutant concentration field exceeds the safety threshold, it is a high-risk area. When the diffusion path involves ecologically sensitive areas or areas with intensive economic activity, it is also a high-risk area. The safety threshold is set at 0.1 mg / L. Based on the definition of high-risk areas, high-concentration distribution areas are extracted from the pollutant path, annotated in grid form, and generated into geographic coordinates.

[0089] Step S53: Compare the pollutant concentration field obtained by monitoring with the pollutant concentration field obtained by pollutant model simulation, specifically:

[0090] ∈(x,y,z,t)=C 监测 (x,y,z,t)-C 模拟 (x,y,z,t)

[0091] Where: ∈(x,y,z,t) is the error distribution, C 监测 (x, y, z, t) is the real-time monitoring concentration field, C 模拟 (x, y, z, t) is the pollutant concentration field calculated by the model, (x, y, z) is the spatial coordinate, and t is the time.

[0092] When the error is less than the threshold, the verification is passed. If it exceeds the threshold, iterative adjustments are made. Based on the verified pollutant path, a pollutant diffusion path map is drawn to show the dynamic distribution of pollutants, including a concentration value line graph to represent the concentration gradient change and different colors to mark high-risk areas and low-risk areas. Key areas are highlighted on the diffusion path map, including the source of pollutants and the scope of influence, high-concentration areas and diffusion boundaries, and the action paths of dynamic factors of wind fields and tides, and a risk assessment report is generated. The risk assessment report includes the specific location and area of ​​high-risk areas, the scope of influence and potential risks of the diffusion path, and recommended control measures. The control measures include: setting up isolation zones or cleaning up pollutants.

[0093] In the specific implementation process of step S6, control measures are set based on the obtained risk assessment report and pollutant diffusion path map, and the sensor network is used to monitor the changes in the pollutant concentration field. The changes in pollutant concentration in high-risk areas after cleaning are recorded. The control effect is evaluated based on the obtained risk assessment report and pollutant diffusion path map. A decision support report is generated based on the risk assessment report, pollutant diffusion path map, multi-scale feature description and pollutant concentration field, specifically including:

[0094] Step S61: setting control measures based on the obtained risk assessment report and pollutant diffusion path map;

[0095] The control measures include the establishment of isolation zones, pollutant cleanup measures, chemical neutralization measures, and bioremediation measures. The isolation zones are set up by combining the dynamic field vectors and setting up physical isolation zones at the boundaries of the diffusion path to limit the further spread of pollutants.

[0096] Pollutant cleanup measures include adsorption and mechanical salvage. Adsorption involves using adsorbents to remove pollutants in high-concentration areas based on a pollutant profile matrix. Mechanical salvage involves using mechanical equipment to clean surface pollutants based on a pollutant diffusion pathway map. Chemical neutralization involves selecting neutralizers based on chemical reaction models. Neutralizers include acidic or alkaline solutions to adjust pH and chelating agents to reduce the activity of heavy metal pollutants. Bioremediation involves the introduction of microbial agents into areas of long-term diffusion based on a pollutant profile matrix.

[0097] Step S62: Using the sensor network to monitor changes in the pollutant concentration field, and record changes in pollutant concentrations in high-risk areas after cleaning;

[0098] Adjust control measures based on the comparison error between the monitored concentration field and the pollutant concentration field, including updating the isolation zone location setting according to the pollutant diffusion path map, increasing the cleaning frequency of high-concentration areas, and adjusting the dosage to match the latest chemical reaction rate.

[0099] Step S63: Evaluate the treatment effect based on the obtained risk assessment report and pollutant diffusion path map.

[0100] The evaluation of treatment effect includes concentration reduction rate and diffusion area reduction rate.

[0101] The concentration reduction rate is specifically expressed as:

[0102]

[0103] Where: Concentration reduction rate is the percentage of pollutant concentration reduction, which is used to evaluate the treatment effect, C 监测 is the initial concentration of the pollutant before treatment, and ×100% means converting the result into percentage.

[0104] The diffusion area reduction rate is specifically expressed as:

[0105]

[0106] Where: Diffusion area reduction rate is the percentage of diffusion area reduction, which is used to evaluate the diffusion control effect, A 初始 is the initial diffusion area of ​​pollutants before treatment, A 监测 It is the pollutant diffusion area after treatment.

[0107] If the concentration reduction rate is >80% and the diffusion area reduction rate is >70%, the treatment effect is judged to be qualified. If the effect does not meet the standards, the treatment plan will be adjusted and continued to be implemented.

[0108] Step S64: generating a decision support report based on the risk assessment report, the pollutant diffusion path map, the multi-scale feature description, and the pollutant concentration field;

[0109] Generate a decision support report that includes concentration changes before and after treatment in high-risk areas, treatment results of diffusion paths and impacts on ecologically sensitive areas, and proposes further optimized treatment measures and treatment effect evaluation results for areas that do not meet standards.

[0110] The above description is merely a preferred embodiment of the present invention and does not constitute any other form of limitation to the present invention. Any person skilled in the art may utilize the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes for application in other fields. However, any simple modification, equivalent change, and modification of the above embodiments made in accordance with the technical essence of the present invention without departing from the technical solution of the present invention shall still fall within the scope of protection of the technical solution of the present invention.

Claims

1. A pollutant diffusion simulation and assessment method for marine environmental protection, characterized in that: The steps include: Step S1: Acquire multi-source data and perform pre-processing operations on it to generate a database; The multi-source data includes dynamic factor data, physical and chemical condition data and pollutant characteristic data, and the pre-processed multi-source data generates a dynamic factor data matrix, a physical and chemical condition matrix and a pollutant characteristic matrix respectively; Step S2: performing calculations based on the pollutant model and the chemical reaction model to obtain a pollutant concentration field; The pollutant model is configured based on the preprocessed multi-source data, and the dynamic factor is initialized. The data after dynamic factor initialization is combined with the pollutant characteristic matrix to construct a chemical reaction model, thereby obtaining the pollutant concentration field; Step S3, generating multi-scale feature description; The pollutant concentration field is decomposed into short-term diffusion and long-term diffusion, and the short-term diffusion characteristics and long-term diffusion characteristics are obtained accordingly. Then, the microscopic and macroscopic characteristics of the pollutant concentration field are extracted, and the short-term diffusion characteristics and long-term diffusion characteristics are combined to generate a multi-scale feature description. The specific steps include: Step S31: Decompose the pollutant concentration field into short-term diffusion and long-term diffusion, specifically: ； Where: is the pollutant concentration field, representing any time and spatial location The distribution of pollutant concentrations, It is a short-term diffusion characteristic that describes the instantaneous fluctuation of pollutants in a short period of time. It is a long-term diffusion characteristic, which indicates the average concentration distribution trend of pollutants over a long time scale; Step S32: Perform frequency domain analysis on the time signal of the pollutant concentration field using fast Fourier transform to extract high-frequency components, calculate the long-term variation trend of the pollutant concentration field using time-weighted average method, and obtain microscopic characteristics; use wavelet transform to decompose the pollutant concentration field into multiple spatial resolution characteristics to obtain macroscopic characteristics; Step S33: Combine the short-term diffusion features, long-term diffusion features, microscopic features, and macroscopic features to generate a multi-scale feature description, specifically: ； Where: It is a multi-scale feature description. It is The characteristic components of spatial scales, is a short-term diffusion characteristic. is a long-term diffusion characteristic. Represents the superposition of all spatial scales; Step S4, optimizing the pollutant model based on multi-scale feature description; Monitor pollutant distribution through sensor networks and obtain pollutant concentration fields. By comparing monitoring data with model calculation results, identify errors in the pollutant diffusion model and dynamically optimize the pollutant model by combining multi-scale feature descriptions. Step S5: Use the optimized pollutant model to simulate the diffusion path of pollutants, compare the monitored pollutant concentration field with the simulated pollutant concentration field, draw a pollutant diffusion path map, highlight high-risk areas on the diffusion path map, and generate a risk assessment report.

2. A pollutant diffusion simulation and assessment method for marine environmental protection according to claim 1, characterized in that: In step S2, the data after the dynamic factor is initialized include ocean current speed and direction, tidal height and period, wind field intensity and direction, temperature, salinity and pH value, which is achieved in the following way: Step S21: Using the multidimensional dynamics of pollutants as the mathematical basis of the pollutant model, specifically: ； Where: is the pollutant concentration, It's time, is the convection term, is the gradient operator, is the dynamic field vector, is the diffusion term, is the diffusion coefficient, is the Laplace operator, is the chemical reaction term, is the temperature, is salinity, It is pH; Step S22: Based on the grid structure of the pollutant model, the dynamic factors are assigned and calculated to establish a spatiotemporal continuous dynamic field. Under the action of the dynamic field, the pollutant concentration field is calculated in combination with the chemical reaction model; (1) Extract ocean current speed and direction, tidal height and period, and wind field intensity and direction from the dynamic factor data matrix; extract temperature, salinity, and pH value from the physical and chemical condition matrix; Temperature is used to affect diffusion coefficients and reaction rates; (2) Based on the extracted ocean current speed and direction, tidal height and period, wind field intensity and direction, temperature, salinity and pH value, and combined with the pollutant characteristic matrix, define the chemical reaction model; ； Where: is the chemical reaction rate constant, is the pollutant concentration; Extract the initial input concentration from the pollutant characteristic matrix and mark it as the initial distribution of the pollutant. Based on the defined chemical reaction model, set the initial concentration value of each cell in the grid area and combine it with the diffusion boundary condition to generate the initial concentration field. Calculations are performed based on the configured pollutant model and chemical reaction model to obtain the pollutant concentration field; The grid structure of the pollutant model is used to establish a spatiotemporal continuous dynamic field, allowing pollutants to be transported and diffused under the influence of ocean currents, tides, and wind field dynamic factors, thereby obtaining a pollutant concentration field: ； Where: is the pollutant concentration field, representing any time and spatial location The distribution of pollutant concentrations, is the rate of change of the concentration field with respect to time, is the concentration distribution under the action of the dynamic field, is a chemical reaction term that describes the transformation behavior of pollutants under physical and chemical conditions. is the temperature, is salinity, It is concentration, the core variable of chemical reactions.

3. The pollutant diffusion simulation and assessment method for marine environmental protection according to claim 1, characterized in that: The step S4 is specifically implemented in the following manner: Step S41: Extract the rapid transport path of pollutants based on the obtained short-term diffusion characteristics and dynamically adjust the pollutant concentration field; calculate the diffusion rate of pollutants in different grid cells based on the long-term diffusion characteristics and correct the diffusion coefficient; and adjust the reaction rate constant based on the physical and chemical condition data; Step S42: Monitor the pollutant distribution through the sensor network and obtain the monitored pollutant concentration field, and calculate the error by comparing the monitored pollutant concentration field with the pollutant concentration field obtained by the pollutant model operation; through the comparative calculation, if the error exceeds the set threshold, iterative adjustment is triggered; the iterative adjustment includes correcting the flow velocity direction of the dynamic field, updating the regional distribution of the diffusion coefficient, and correcting the conversion rate of the chemical reaction term.

4. A pollutant diffusion simulation and assessment method for marine environmental protection according to claim 3, characterized in that: The step S5 specifically includes the following steps: Step S51: simulating the diffusion path of pollutants using the optimized pollutant model to obtain the pollutant path; Extract the diffusion path of pollutants based on the pollutant path, mark the points where the concentration value reaches the warning threshold, track its diffusion direction and speed, and generate a pollutant diffusion path map to show the diffusion trajectory of pollutants from the source to the surrounding area; Step S52: defining high-risk areas based on the obtained pollutant diffusion path map; When the pollutant concentration in the pollutant concentration field exceeds the safety threshold, it is a high-risk area. When the diffusion path involves ecologically sensitive areas or areas with intensive economic activities, it is a high-risk area. The safety threshold is set at 0.1 mg / L; Based on the definition of high-risk areas, high-concentration distribution areas are extracted from the pollutant paths, marked in grid form, and geographic coordinates are generated; Step S53: comparing the pollutant concentration field based on the monitoring with the pollutant concentration field calculated by the optimized pollutant model; When the error is less than the threshold, the verification is passed. If it exceeds the threshold, iterative adjustment is performed. Based on the verified pollutant path, a pollutant diffusion path map is drawn to display the dynamic distribution of pollutants, including a concentration value line graph to represent the concentration gradient change and high-risk areas and low-risk areas marked with different colors. Key areas are highlighted on the diffusion path map, including the source of pollutants and the scope of influence, high-concentration areas and diffusion boundaries, and the action paths of the dynamic factors of wind fields and tides, and a risk assessment report is generated.

5. The pollutant diffusion simulation and assessment method for marine environmental protection according to claim 1 is characterized in that: The step S5 also includes the step of formulating treatment measures, specifically: Step S6: Set control measures based on the risk assessment report and the pollutant diffusion path map, evaluate the control effect in combination with the control effect feedback, adjust the control measures, and finally generate a decision support report.

6. A pollutant diffusion simulation and assessment method for marine environmental protection according to claim 5, characterized in that: In step S6: (1) The control measures include the establishment of isolation zones, pollutant cleanup measures, chemical neutralization measures and bioremediation measures; (2) The treatment effect evaluation indicators include the concentration reduction rate and the diffusion area reduction rate; if the concentration reduction rate is greater than 0% and the diffusion area reduction rate is greater than 0%, the treatment effect is judged to be qualified; if the effect does not meet the standards, the treatment plan is adjusted and continued to be implemented; (3) Adjusting control measures based on the comparison error between the monitored concentration field and the pollutant concentration field, including updating the location of isolation zones based on the pollutant diffusion path map, increasing the cleaning frequency of high-concentration areas, and adjusting the dosage to match the latest chemical reaction rate; (4) The decision support report includes the concentration changes before and after the treatment of high-risk areas, the treatment results of the diffusion path and the impact on ecologically sensitive areas.

7. The pollutant diffusion simulation and assessment method for marine environmental protection according to claim 1 is characterized in that: The preprocessing operations in step S1 include denoising, outlier detection and correction, and data interpolation and reconstruction processes: De-noising uses the Kalman filter algorithm to denoise the collected ocean current speed and direction data to eliminate the environmental noise interference to the sensor; Outlier detection and correction: The box plot method is used to detect outliers, abnormal tide height and wind speed data are marked as abnormal data points, and unreasonable data detected in temperature and salinity distribution are corrected through interpolation correction; Data interpolation and reconstruction uses Lagrange interpolation to fill in gaps in the collected data and reconstruct complete data sets in areas not covered by the monitoring network.

Images