A marine ecological big data analysis method based on marine environment information

Abstract

The application discloses a marine ecological big data analysis method based on marine environment information, relates to the technical field of marine data processing, and comprises the following steps: constructing a diffusion path model based on marine environment data; generating diffusion path weights based on the spatial relationship between the diffusion path model and data collection points; calculating first priority values of data based on the diffusion path weights and obtaining spatial risk propagation characteristics; obtaining time-dimension data by time serialization processing of the diffusion path weights, constructing a space-time joint weight model based on the diffusion path weights and the time-dimension data, obtaining a space-time weight matrix based on the space-time joint weight model, calculating second priority values of data based on the space-time weight matrix and obtaining space-time risk propagation characteristics; extracting features based on marine environment data to obtain a feature vector, obtaining a diffusion model library subset based on the feature vector; and selecting an optimal diffusion model from the diffusion model library subset according to the space-time risk propagation characteristics.

Description

Technical Field

[0001] The present invention relates to the field of marine data processing technology, and in particular to a marine ecological big data analysis method based on marine environmental information. Background Art

[0002] Marine ecosystems are among the most complex and important on Earth. Their health is closely linked to global climate change, biodiversity conservation, the sustainable use of fisheries, and the human environment. In recent years, with the rapid development of ocean observation technology, the ability to obtain marine environmental information and ecological data has exploded, generating a veritable ocean ecosystem of big data. However, effectively utilizing this massive, multi-source, and heterogeneous marine environmental information and ecological big data to deeply understand complex marine ecological processes and achieve accurate ecological assessment, prediction, and management still faces a series of daunting technical challenges.

[0003] Prior art CN119046018B discloses a method for processing and analyzing marine ecological data impacted by the marine environment, comprising: Step 1: determining data collection points in a marine area based on historically collected data; Step 2: performing data analysis on the historically collected marine ecological data to determine individual intervals, then calculating the environmental norm of the marine ecological data based on the data frequency in the individual intervals, and then combining the real-time collected marine ecological data with the environmental norm to calculate the ecological risk value of the data collection point; Step 3: determining the abnormal collection points and the edge collection points, and the straight-line distance between them, and combining this with the ecological risk value of the abnormal collection points to calculate a data priority value, arranging the data priority values ​​to obtain a priority sequence, and then processing the real-time collected marine ecological data according to the priority sequence, thereby improving the response efficiency of risk data and enhancing the overall risk prevention and control capabilities. However, relying solely on the straight-line distance between the abnormal collection points and the edge collection points to calculate the data priority value and construct the priority sequence lacks consideration of dynamic changes in the time dimension. Summary of the Invention

[0004] This application provides a marine ecological big data analysis method based on marine environmental information, which solves the key problems in the existing technology, such as over-simplification of spatial relationship modeling, lack of temporal dynamics, insufficient analysis of risk propagation characteristics, poor model adaptability and delayed passive response, and achieves high-precision, dynamic, and proactive prediction and prevention and control of marine ecological risk management technology.

[0005] This application provides a method for analyzing marine ecological big data based on marine environmental information, including:

[0006] S1: Obtain marine environmental data and construct a diffusion path model based on the marine environmental data; generate diffusion path weights based on the spatial relationship between the diffusion path model and the data collection points; calculate the first priority value of the data based on the diffusion path weights and obtain spatial risk propagation characteristics;

[0007] S2: Obtain time dimension data based on time series processing of diffusion path weights, construct a space-time joint weight model based on the diffusion path weights and time dimension data, obtain a spatiotemporal weight matrix based on the spatiotemporal weight matrix, and calculate the second priority value of the data based on the spatiotemporal weight matrix to obtain the spatiotemporal risk propagation characteristics;

[0008] S3: Extract features based on marine environmental data to obtain feature vectors, and obtain a subset of the diffusion model library based on the feature vectors;

[0009] S4: Select the optimal diffusion model from the diffusion model library subset based on the spatiotemporal risk propagation characteristics, perform spatiotemporal weight correction through the optimal diffusion model and the spatiotemporal weight matrix, and output the optimal decision.

[0010] Furthermore, the marine environment data includes: ocean current data, topographic data, pollution source data, and meteorological data;

[0011] The diffusion path model includes: using a finite element fluid dynamics model to simulate the diffusion trajectory of pollutants under the influence of ocean currents and terrain, and obtaining a diffusion path probability map.

[0012] Furthermore, the diffusion path weight includes: determining an abnormal collection point, predicting the probability and time delay of its pollutants reaching adjacent collection points based on a diffusion path model, and using the prediction results as the diffusion path weight;

[0013] The spatial risk propagation characteristics include: modifying the data priority value formula according to the diffusion path weight to obtain the data first priority value, and analyzing the risk propagation direction, intensity, and key node characteristics from the data first priority value distribution as the spatial risk propagation characteristics.

[0014] Furthermore, the space-time joint weight model includes: generating diffusion path weights of pollutant diffusion paths through a fluid dynamics model, and simultaneously extracting tidal cycles and storm events through time series analysis to construct a space-time joint weight model;

[0015] The spatiotemporal weight matrix includes: performing spatiotemporal convolution fusion on the diffusion path weight and the time dimension data, and optimizing the weight coefficients of each dimension through reinforcement learning, and finally forming a spatiotemporal weight matrix that changes in real time with the environment, driving the calculation of the second priority value of the data.

[0016] Furthermore, the spatiotemporal risk propagation characteristics include:

[0017] The data first priority value formula is modified according to the spatiotemporal weight matrix to obtain the data second priority value. The propagation direction, risk intensity and key node characteristics are analyzed from the data second priority value distribution as the spatiotemporal risk propagation characteristics; the steady-state, turbulent, periodic and composite diffusion modes are identified according to the change law of the data second priority value.

[0018] Furthermore, the characteristic vector includes: vertical velocity variance, tidal phase index, tidal range asymmetry coefficient, sea level height anomaly and velocity mutation sign;

[0019] The diffusion model library subset includes: internal wave diffusion model, tidal asymmetry model and circulation mutation model.

[0020] Furthermore, the optimal diffusion model is to evaluate the explanatory power and predictive fit of each model in the diffusion model library subset for the current spatiotemporal risk characteristics, and select the optimal diffusion model.

[0021] Furthermore, the spatiotemporal weight correction includes:

[0022] The diffusion model is used to predict the probability distribution of pollutant diffusion paths based on the current environmental characteristic vector. The predicted results are compared with the diffusion path weights in the spatiotemporal weight matrix to generate a corrected gradient, and the corrected weight matrix is ​​used to calculate the third priority value of the data.

[0023] Furthermore, the optimal decision includes:

[0024] A collaborative model of pollutant diffusion rate and resource deployment rate is established. The model generates resource scheduling benchmark values ​​through a dynamic matching algorithm, implements decision corrections based on environmental dynamics, and realizes bidirectional coupling through spatiotemporal weight matrix correction to drive resource pre-deployment strategies.

[0025] Furthermore, the resource pre-deployment strategy includes:

[0026] The virtual priority of key nodes is calculated based on the resource scheduling benchmark value and the spatiotemporal weight matrix; protection resources are dispatched to the node with the highest virtual priority before the arrival of pollutants; and the spatiotemporal weight correction gradient is dynamically adjusted through reinforcement learning to achieve dynamic environmental response.

[0027] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0028] By using a fluid dynamics model to construct a diffusion path probability map, and combining the diffusion path weight to correct the calculated data first priority value, the prediction accuracy of spatial risk propagation characteristics is improved; using spatiotemporal convolution to fuse the diffusion path weight and time dimension data, the second priority value of the data after the spatiotemporal weight matrix is ​​optimized is generated, and dynamic analysis of spatiotemporal risk propagation characteristics is achieved; using a random forest classifier to output a subset of the diffusion model library, matching the optimal diffusion model to generate a correction gradient, and dynamic calibration of the spatiotemporal weight matrix is ​​achieved; using a pollutant diffusion-resource deployment collaborative model, virtual priorities are calculated based on the spatiotemporal weight matrix and resource pre-deployment strategies are executed to achieve proactive decision output. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 This is a flow chart of a method for analyzing marine ecological big data based on marine environmental information in an embodiment of the present invention. DETAILED DESCRIPTION

[0030] To facilitate understanding of the present invention, the present application will be described more comprehensively below with reference to the relevant drawings; the drawings show preferred embodiments of the present invention, but the present invention can be implemented in many different forms and is not limited to the embodiments described herein; on the contrary, the purpose of providing these embodiments is to enable a more thorough and comprehensive understanding of the disclosed content of the present invention.

[0031] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this invention pertains; the terms used herein in the specification of the present invention are for the purpose of describing specific embodiments only and are not intended to limit the present invention; the term "and / or" used herein includes any and all combinations of one or more of the associated listed items.

[0032] Example 1: Figure 1 As shown in the figure, a marine ecological big data analysis method based on marine environmental information.

[0033] S1: Obtain marine environmental data and construct a diffusion path model based on the marine environmental data; generate diffusion path weights based on the spatial relationship between the diffusion path model and the data collection points; calculate the first priority value of the data based on the diffusion path weights and obtain spatial risk propagation characteristics;

[0034] Said marine environment data include: ocean current data, topographic data, pollution source data, and meteorological data;

[0035] Specifically, real-time or historical ocean current speed and direction are obtained as ocean current data through buoy monitoring and satellite remote sensing; seabed topography elevation maps are obtained through multi-beam bathymetry systems and public terrain databases, and global seabed elevation data are integrated as terrain data; pollutant types, release amounts, and densities are recorded as pollution source data through on-site monitoring platforms and remote sensing inversion; real-time wind speed and wind direction monitoring and storm event paths are obtained as meteorological data through meteorological buoys and meteorological satellites.

[0036] The diffusion path model includes: using a finite element fluid dynamics model to simulate the diffusion trajectory of pollutants under the influence of ocean currents and terrain, and obtaining a diffusion path probability map.

[0037] Specifically, the diffusion path model is constructed using the fluid dynamics equation, and the formula is:

[0038]

[0039] in, is the rate of change of pollutant concentration over time, The advection of pollutants driven by ocean currents, is the pollutant mixing caused by turbulent diffusion, is the release intensity of the pollution source, C is the pollutant concentration, u is the ocean current velocity field, and D is the diffusion coefficient.

[0040] Convert the seabed terrain elevation map into a terrain boundary constraint function, and add the velocity attenuation coefficient to the ridge area:

[0041]

[0042] in, is the terrain resistance parameter, is the relative height of the ridge, is the velocity attenuation coefficient;

[0043] Add diffusion suppression coefficient to the shallow area:

[0044]

[0045] Where H is the actual water depth, is the critical shallow water threshold, is the suppression intensity coefficient, is the diffusion inhibition coefficient.

[0046] The velocity attenuation coefficient and diffusion suppression coefficient of the terrain boundary constraint function are embedded in the fluid dynamics equation, and the modified formula is:

[0047]

[0048] in, for , for .

[0049] Monte Carlo simulation is used to generate a diffusion path probability map, and the particle motion trajectory is corrected according to the flow velocity attenuation coefficient and the diffusion inhibition coefficient.

[0050] The diffusion path weight includes: determining an abnormal collection point, predicting the probability and time delay of its pollutants reaching the adjacent collection points based on the diffusion path model, and using the prediction results as the diffusion path weight;

[0051] Specifically, the average value of marine ecological data is used as the normal environmental value. The abnormal collection point refers to the area with marine environmental risks or problems calculated by real-time marine ecological data and normal environmental values. For each pollutant collection point, Monte Carlo simulation is performed based on the diffusion path model; the pollution source data, ocean current velocity field, and diffusion coefficient As input data, the probability of reaching the adjacent collection point is obtained With time delay , normalize the obtained probability and time delay to the same dimension [0,1], and calculate to obtain the diffusion path weight:

[0052]

[0053] in, is the diffusion path weight.

[0054] The spatial risk propagation characteristics include: modifying the data priority value formula according to the diffusion path weight to obtain the data first priority value, and analyzing the risk propagation direction, intensity, and key node characteristics from the data priority value distribution as the spatial risk propagation characteristics.

[0055] Specifically, the data priority value calculation is optimized based on the diffusion path weight:

[0056]

[0057] in, is the data first priority value, is the ecological risk value, is the diffusion path weight.

[0058] Calculate each collection point The spatial gradient vector of the value, the high gradient direction indicates the main diffusion direction of the pollutant, and the risk propagation path is identified by the spatial gradient distribution of the first priority value of the data as the risk propagation direction feature; The numerical value quantifies the risk level. The higher the value, the greater the probability and urgency of the location being affected by pollutants, which is used as an intensity feature; identification The extreme points in the value distribution are used as key node features, where The local peak point of the value is the pollution source node, The area with sudden value increase is the sensitive node, and the transfer node is located at the intersection of multiple diffusion paths. The rate of change of value is significant.

[0059] The technical solutions in the above embodiments of the present application have at least the following technical effects or advantages:

[0060] This application uses fluid dynamics equations to quantify the impact of complex marine environments on pollutant diffusion. Monte Carlo simulation generates probabilistic diffusion paths to improve prediction accuracy. Diffusion path weights are defined, combining pollutant arrival probability with time delay, and dynamically correcting data priority values ​​to make risk assessments more aligned with the actual transmission process. The distribution of data priority values ​​is analyzed to identify direction, intensity, and key nodes, forming a structured spatial risk profile to guide precise prevention and control.

[0061] Example 2: In Example 1, the impact of the marine environment on pollutant diffusion was quantified through a fluid dynamics model, combined with Monte Carlo simulation to generate probabilistic diffusion paths, and spatial risk assessment was modified based on dynamic weights to achieve accurate analysis of the direction, intensity, and key nodes of risk transmission. However, the data priority value was calculated based solely on the current diffusion path weight, without considering the temporal variation of the marine environment. Furthermore, the data priority value did not incorporate time series data into the weight calculation, making it impossible to predict the evolution of risk over time. This example further improves on Example 1.

[0062] S2: Obtain time dimension data based on time series processing of diffusion path weights, construct a space-time joint weight model based on the diffusion path weights and time dimension data, obtain a spatiotemporal weight matrix based on the spatiotemporal weight matrix, and calculate the second priority value of the data based on the spatiotemporal weight matrix to obtain the spatiotemporal risk propagation characteristics;

[0063] Specifically, based on the dynamic characteristics of the ocean environment, time series processing tracks wind speed and tide data at the minute level to generate time dimension data in real time. The time dimension data includes a time correction coefficient and a terrain obstruction coefficient. The time correction coefficient increases linearly with increasing wind speed, capturing the accelerating effect of storms on pollutant diffusion. Its formula is:

[0064]

[0065] in, is the time correction coefficient, 1 is the basic diffusion coefficient benchmark, 0.1 is the diffusion acceleration slope, which means that for every 1m / s increase in wind speed, the pollutant diffusion speed increases by 10%, F is the wind speed input by real-time meteorological data, and 10 is the critical value of ocean fluid mechanics. When the wind speed is greater than 10m / s, the sea surface turbulence is significantly enhanced. For the effective wind speed increment, only the wind speed portion exceeding the threshold (10 m / s) is calculated to avoid invalid disturbances in the low wind speed segment.

[0066] The topographic resistance coefficient increases exponentially with decreasing tide level, quantifying the increased diffusion resistance caused by the exposure of the seabed topography during low tide. Its formula is:

[0067]

[0068] in, is the terrain obstruction coefficient, is the terrain complexity index, and its value range is greater than 0, such as 0 in flat areas and 2 in coral reef areas. k is the terrain sensitivity coefficient, and its default value is 0.8.

[0069] The space-time joint weight model includes: generating diffusion path weights of pollutant diffusion paths through a fluid dynamics model, and extracting tidal cycles and storm events through time series analysis to construct a space-time joint weight model;

[0070] Specifically, the time dimension data is added to the original diffusion path weight to establish a space-time joint weight model:

[0071]

[0072] in, is the joint space-time weight.

[0073] The spatiotemporal weight matrix includes: performing spatiotemporal convolution fusion on the diffusion path weight and the time dimension data, and optimizing the weight coefficients of each dimension through reinforcement learning, and finally forming a spatiotemporal weight matrix that changes in real time with the environment, driving the calculation of the second priority value of the data.

[0074] Specifically, the diffusion path weight and time dimension data are fused into spatiotemporal weight. The fusion formula is:

[0075]

[0076] in, is the spatiotemporal weight, representing the final risk weight of the location (x, y) at time t, is the spatial diffusion weight, is the spatiotemporal convolution operator, is the wind speed influence weight, is the terrain influence weight, and The sum of the values ​​is 1. is the latitude and longitude of the ocean data collection point, and t is the timestamp of data collection.

[0077] Will 、 ,..., As a spatiotemporal weight matrix , further optimize the data first priority value:

[0078]

[0079] in, is the optimized data priority value, is the dynamic adjustment coefficient, and its value range is [1,2].

[0080] The spatiotemporal risk propagation characteristics include: modifying the data first priority value formula according to the spatiotemporal weight matrix to obtain the data second priority value, analyzing the propagation direction, risk intensity and key node characteristics from the data second priority value distribution as the spatiotemporal risk propagation characteristics; identifying steady-state, turbulent, periodic and composite diffusion modes according to the change law of the data second priority value.

[0081] Specifically, through the data second priority value The spatial gradient distribution identifies the main diffusion direction of pollutants as a propagation direction feature; The value size quantifies the risk level and urgency as the risk intensity characteristics, according to The values ​​are divided into three risk levels, <5 is low risk, 5≤ <8 is medium risk, ≥8 is high risk; the pollution source node ( local maximum), sensitive nodes ( Burst rate>20% / hour), transit node ( The change rate is >1.5 / km) as the key node feature.

[0082] Through analysis The spatial distribution pattern and time dimension fluctuation law of the scatterplot are matched with steady-state, turbulent, periodic and composite diffusion patterns; after identifying the pattern, the resource pre-deployment strategy is dynamically adjusted, and the spatiotemporal weight matrix is ​​reversely optimized. parameters to improve the subsequent prediction accuracy.

[0083] The technical solutions in the above embodiments of the present application have at least the following technical effects or advantages:

[0084] This application generates time dimension data in real time by tracking wind speed and tidal data at the minute level, and constructs a space-time joint weight model in combination with the diffusion path weight; the diffusion path weight and the time dimension data are fused through spatiotemporal convolution, and the weight coefficients of each dimension are optimized through reinforcement learning to form a spatiotemporal weight matrix that changes in real time with the environment; the second priority value of the calculated data is optimized based on the spatiotemporal weight matrix to obtain the spatiotemporal risk propagation characteristics, analyze the propagation direction, risk intensity and key node characteristics, and identify steady-state, turbulent, periodic and composite diffusion modes according to the change law of the second priority value of the data, so as to achieve improved dynamic prediction accuracy.

[0085] Example 3: Example 2 constructs a dynamic weight matrix through spatiotemporal convolution fusion , realizing the prediction of risk spatiotemporal evolution, but the weight matrix relies on historical data training and cannot adapt to sudden environmental changes, and is not associated with physical mechanisms, resulting in prediction deviation and decision lag. This embodiment further improves the second embodiment.

[0086] S3: Extract features based on marine environmental data to obtain feature vectors, and obtain a subset of the diffusion model library based on the feature vectors;

[0087] The characteristic vectors include: vertical velocity variance, tidal phase index, tidal range asymmetry coefficient, sea level height anomaly and velocity mutation sign;

[0088] Specifically, the vertical velocity variance is a statistic that characterizes the vertical mixing intensity of the water body and reflects the diffusion capacity of pollutants in the vertical direction; the tidal phase index quantifies the time position in the tidal cycle and is related to the periodic influence of tidal dynamics on diffusion; the tidal range asymmetry coefficient characterizes the nonlinear difference in the duration and intensity of high and low tides and drives the net transport of pollutants; the sea surface height anomaly is the deviation of the sea surface height inverted by satellite remote sensing from the climatological mean; the velocity mutation mark is used to identify the instantaneous sudden change in ocean current velocity.

[0089] S4: Select the optimal diffusion model from the diffusion model library subset based on the spatiotemporal risk propagation characteristics, perform spatiotemporal weight correction through the optimal diffusion model and the spatiotemporal weight matrix, and output the optimal decision.

[0090] The diffusion model library subset includes: internal wave diffusion model, tidal asymmetry model and circulation mutation model.

[0091] Specifically, the diffusion model library subset is a feature vector that is dynamically filtered through a random forest classifier to identify three types of pollutant diffusion physical models. The output is the applicable probability of the three diffusion models. When the probability of a model exceeds a threshold (>0.7), the model is activated and added to the subset to modify the spatiotemporal weight matrix and optimize decision making. If the probabilities of all models are below the threshold, the basic fluid dynamics model is used by default.

[0092] The optimal diffusion model is to evaluate the explanatory power and predictive fit of each model in the diffusion model library subset for the current spatiotemporal risk characteristics, and select the optimal diffusion model.

[0093] Specifically, when the vertical flow velocity variance is significant (>0.05m² / s²) and the sea level height anomaly is positive (>0.1m), the internal wave diffusion model is adopted; when the absolute value of the tidal range asymmetry coefficient is greater than 0.3, the tidal asymmetry model is adopted; when the flow velocity mutation flag is 1, the circulation mutation model is adopted; the circulation mutation model has the highest priority because it involves emergencies, followed by the internal wave diffusion model and the tidal asymmetry model, to ensure that the system responds to the most urgent risks first.

[0094] The spatiotemporal weight correction includes: using a diffusion model to predict the probability distribution of pollutant diffusion paths based on the current environmental characteristic vector, comparing the prediction results with the diffusion path weights in the spatiotemporal weight matrix, generating a correction gradient, and using the corrected weight matrix to calculate the third priority value of the data.

[0095] Specifically, the optimal diffusion model that matches the current environment is used to predict the probability distribution map of the pollutant diffusion path based on the real-time feature vector. The formula is:

[0096]

[0097] in, The model predicts the arrival probability of pollutants, is the diffusion model, is the feature vector.

[0098] The model prediction results and the spatiotemporal weight matrix Diffusion path weight in Perform a difference comparison:

[0099]

[0100] in, Reflects the deviation between the model prediction and the actual weight, generating a correction gradient.

[0101] Update the diffusion path weights by correcting the gradient:

[0102]

[0103] in, is the updated diffusion path weight, is the learning rate, with a range of [0.05, 0.2], which controls the correction amplitude. Resubstitute the spatiotemporal weight matrix , generate the corrected matrix .

[0104] Based on the modified Calculate the third priority value of data:

[0105]

[0106] in, The third priority value for data, The integration of physical model prediction results makes the analyzed spatiotemporal risk propagation characteristics closer to the real ocean dynamics process.

[0107] The technical solutions in the above embodiments of the present application have at least the following technical effects or advantages:

[0108] This application uses a random forest classifier to generate a subset of the diffusion model library, including the internal wave diffusion model, the tidal asymmetry model, and the circulation mutation model, and dynamically matches the optimal diffusion model based on the physical threshold, among which the circulation mutation model has the highest priority;

[0109] The optimal diffusion model is used to predict the probability of pollutant diffusion paths, which is compared with the diffusion path weights in the spatiotemporal weight matrix to generate a corrected gradient. The weights are updated through the learning rate η, the spatiotemporal weight matrix is ​​reconstructed, and the third priority value of the data is calculated. By coupling the physical model with real-time data, the risk prediction accuracy and decision-making timeliness in sudden marine environments are significantly improved.

[0110] Example 4: In Example 3, a correction gradient is generated by the difference between the model prediction results and the spatiotemporal weight matrix, and the diffusion path weight is updated in real time through the dynamic learning rate, so that the risk prediction can adapt to sudden environmental changes and reduce the prediction error in the historical data failure scenario by 30%-50%. However, its optimization only stays at the risk feature analysis level, and the association between the prediction results and resource scheduling decisions is not established, resulting in the inability to convert high-precision predictions into proactive prevention and control actions. This example further improves Example 3.

[0111] Step S4 also includes: outputting the optimal decision and providing proactive technical support.

[0112] The optimal decision includes: establishing a collaborative model of pollutant diffusion speed and resource deployment speed. The model generates a resource scheduling benchmark value through a dynamic matching algorithm, implements decision correction based on environmental dynamics, realizes bidirectional coupling through spatiotemporal weight matrix correction, and drives resource pre-deployment strategy.

[0113] Specifically, the resource scheduling benchmark value is calculated based on the matching relationship between the pollutant diffusion speed and the resource deployment speed:

[0114]

[0115] in, is the resource scheduling benchmark value, is the diffusion speed of the dynamic prediction of spatiotemporal risk propagation characteristics, The speed of resource deployment, is the environmental dynamic adjustment coefficient; and Both are spatiotemporal weight matrices Real-time output, Scheduling database for resources from pre-defined sources.

[0116] The resource pre-deployment strategy includes: calculating the virtual priority of key nodes based on the resource scheduling benchmark value and the spatiotemporal weight matrix; scheduling protective resources to the nodes with the highest virtual priority before the arrival of pollutants; and dynamically adjusting the spatiotemporal weight correction gradient through reinforcement learning to achieve dynamic environmental response.

[0117] Specifically, the resource scheduling benchmark value and the spatiotemporal weight matrix are used as input data to calculate the virtual priority:

[0118]

[0119] in, is the virtual priority value, is the node sensitivity coefficient, which is 1.5 for key nodes and 0.8 for edge nodes. The top 10% nodes are regarded as key nodes, and resources are deployed to them before pollutants arrive. The highest node.

[0120] Update the learning rate η through resource deployment effect feedback:

[0121]

[0122] in, is the learning rate, is the actual pollution reduction rate, derived from the detection data of the deployed sensor network, To predict the reduction rate, simulation prediction results from the optimal diffusion model are used.

[0123] Changing the local environment through resource deployment feeds back to the spatiotemporal weight matrix , triggering a new round of weight adjustments. This has achieved an intelligent closed loop of dynamic environmental perception, real-time model iteration, and rolling decision optimization, significantly improving the initiative and accuracy of marine pollution risk prevention and control.

[0124] The technical solutions in the above embodiments of the present application have at least the following technical effects or advantages:

[0125] This application generates a resource scheduling benchmark value by establishing a collaborative model of pollutant diffusion speed and resource deployment speed, and calculates the virtual priority value of key nodes based on the corrected data third priority value and the resource scheduling benchmark value, driving the resource pre-deployment strategy, and scheduling protective resources to the node with the highest virtual priority before the arrival of pollutants; at the same time, it dynamically optimizes the system through a reinforcement learning mechanism: using the feedback comparison between the actual pollution reduction rate and the predicted reduction rate, the learning rate of spatiotemporal weight correction is adaptively adjusted to form a closed loop of environmental dynamic perception-model prediction-resource pre-deployment-effect feedback-weight iteration, transforming high-precision risk prediction into proactive prevention and control actions, shortening resource response time by more than 50%, and increasing the prevention and control coverage rate of key nodes to 95%, significantly enhancing the system's adaptability to sudden marine environmental events and decision-making accuracy.

[0126] The foregoing description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Various modifications and variations are readily apparent to those skilled in the art. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention shall be included within the scope of protection of the present invention.

Claims

1. A marine ecological big data analysis method based on marine environmental information, characterized in that: include: S1: Acquire marine environmental data and build a diffusion path model based on the marine environmental data; Generate diffusion path weights based on the spatial relationship between the diffusion path model and the data collection points; Calculate the first priority value of the data based on the diffusion path weight and obtain the spatial risk propagation characteristics; The diffusion path weight includes: determining an abnormal collection point, predicting the probability and time delay of its pollutants reaching the adjacent collection points based on the diffusion path model, and using the prediction results as the diffusion path weight; S2: Obtain time dimension data based on time series processing of diffusion path weights, construct a space-time joint weight model based on the diffusion path weights and time dimension data, obtain a spatiotemporal weight matrix based on the spatiotemporal weight matrix, and calculate the second priority value of the data based on the spatiotemporal weight matrix to obtain the spatiotemporal risk propagation characteristics; The space-time joint weight model includes: generating diffusion path weights of pollutant diffusion paths through a fluid dynamics model, and extracting tidal cycles and storm events through time series analysis to construct a space-time joint weight model; The spatiotemporal weight matrix includes: performing spatiotemporal convolution fusion on the diffusion path weight and the time dimension data, and optimizing the weight coefficients of each dimension through reinforcement learning, and finally forming a spatiotemporal weight matrix that changes in real time with the environment; S3: Extract features based on marine environmental data to obtain feature vectors, and obtain a subset of the diffusion model library based on the feature vectors; S4: Select the optimal diffusion model from the diffusion model library subset based on the spatiotemporal risk propagation characteristics, perform spatiotemporal weight correction through the optimal diffusion model and the spatiotemporal weight matrix, and output the optimal decision.

2. A method for analyzing marine ecological big data based on marine environmental information according to claim 1, characterized in that: Said marine environment data include: ocean current data, topographic data, pollution source data, and meteorological data; The diffusion path model includes: using a finite element fluid dynamics model to simulate the diffusion trajectory of pollutants under the influence of ocean currents and terrain, and obtaining a diffusion path probability map.

3. The method for analyzing marine ecological big data based on marine environmental information according to claim 1, characterized in that: The spatial risk propagation characteristics include: modifying the data priority value formula according to the diffusion path weight to obtain the data first priority value, and analyzing the risk propagation direction, intensity, and key node characteristics from the data first priority value distribution as the spatial risk propagation characteristics.

4. The method for analyzing marine ecological big data based on marine environmental information according to claim 1, wherein: The spatiotemporal risk propagation characteristics include: The data first priority value formula is modified according to the spatiotemporal weight matrix to obtain the data second priority value. The propagation direction, risk intensity and key node characteristics are analyzed from the data second priority value distribution as the spatiotemporal risk propagation characteristics; the steady-state, turbulent, periodic and composite diffusion modes are identified according to the change law of the data second priority value.

5. The method for analyzing marine ecological big data based on marine environmental information according to claim 1, characterized in that: The characteristic vectors include: vertical velocity variance, tidal phase index, tidal range asymmetry coefficient, sea level height anomaly and velocity mutation sign; The diffusion model library subset includes: internal wave diffusion model, tidal asymmetry model and circulation mutation model.

6. A method for analyzing marine ecological big data based on marine environmental information according to claim 1, characterized in that: The optimal diffusion model is to evaluate the explanatory power and predictive fit of each model in the diffusion model library subset for the current spatiotemporal risk characteristics, and select the optimal diffusion model.

7. The method for analyzing marine ecological big data based on marine environmental information according to claim 1, characterized in that: The spatiotemporal weight modification includes: The diffusion model is used to predict the probability distribution of pollutant diffusion paths based on the current environmental characteristic vector. The predicted results are compared with the diffusion path weights in the spatiotemporal weight matrix to generate a corrected gradient, and the corrected weight matrix is ​​used to calculate the third priority value of the data.

8. The method for analyzing marine ecological big data based on marine environmental information according to claim 1, characterized in that: The optimal decision includes: A collaborative model of pollutant diffusion rate and resource deployment rate is established. The model generates resource scheduling benchmark values ​​through a dynamic matching algorithm, implements decision corrections based on environmental dynamics, and realizes bidirectional coupling through spatiotemporal weight matrix correction to drive resource pre-deployment strategies.

9. A method for analyzing marine ecological big data based on marine environmental information according to claim 8, characterized in that: The resource pre-deployment strategy includes: The virtual priority of key nodes is calculated based on the resource scheduling benchmark value and the spatiotemporal weight matrix; protection resources are dispatched to the node with the highest virtual priority before the arrival of pollutants; and the spatiotemporal weight correction gradient is dynamically adjusted through reinforcement learning to achieve dynamic environmental response.

Images