Microplastic into-sea transportation simulation method based on plume model and artificial intelligence model

Abstract

The invention belongs to the field of environmental pollution control, and discloses a micro-plastic sea-entering transportation simulation method based on a plume model and an artificial intelligence model, and the method comprises the steps: collecting micro-plastic attribute data, environment data and micro-plastic historical transportation data of an estuary region; constructing a plume model; inputting the micro-plastic attribute data and the environment data into the plume model to obtain a preliminary transportation result; training an artificial intelligence model; inputting the preliminary transportation result into the trained artificial intelligence model to obtain micro-plastic transportation process prediction data, and iteratively optimizing key parameters of the plume model based on the micro-plastic transportation process prediction data and micro-plastic transportation process real data to obtain an integrated optimized plume model and artificial intelligence model; and inputting environment data and micro-plastic attribute data which are acquired in real time into the plume model and the artificial intelligence model which are integrated and optimized to obtain a micro-plastic transportation simulation result. According to the invention, accurate transportation simulation of the micro-plastic in the estuary area can be realized.

Description

Technical Field

[0001] The present invention belongs to the field of environmental pollution control, and particularly relates to a method for simulating the transport of microplastics into the sea based on a plume model and an artificial intelligence model. Background Art

[0002] As a new type of pollutant, microplastics have a certain impact on the marine ecosystem and biological health. The accurate simulation of the transport characteristics and the input flux of microplastics into the sea is the basis for evaluating its pollution impact, predicting the pollution trend, and designing pollution prevention and control measures.

[0003] Currently, numerical models can be used to simulate the transport of microplastics. However, since the behavior of microplastics in the environment is affected by various factors, such as the influence of tides, winds, etc., the current mathematical models are difficult to capture these complex dynamic changes, resulting in a relatively low accuracy of the simulation results of the transport of microplastics. Summary of the Invention

[0004] The purpose of the present invention is to solve the technical problem of the relatively low accuracy of the simulation results of the transport of microplastics in the prior art. By combining the plume numerical model with modern artificial intelligence technology, the accurate transport simulation of microplastics in the estuary area is realized, providing effective technical support for the control of microplastic pollution and ecological risk assessment.

[0005] In a first aspect, an embodiment of the present invention provides a method for simulating the transport of microplastics into the sea based on a plume model and an artificial intelligence model. The method includes: Collecting the microplastic attribute data, environmental data, and historical transport data of microplastics in the estuary area; Constructing a plume model based on the hydrological conditions and microplastic characteristics in the estuary area, where the plume model takes into account the influence of environmental factors on the transport process of microplastics; Inputting the microplastic attribute data and the environmental data into the plume model, and simulating the transport process of microplastics by means of numerical calculation methods to obtain preliminary transport results; Training an artificial intelligence model based on the environmental data and the historical transport data of microplastics, and the trained artificial intelligence model can predict the transport process of microplastics; Inputting the preliminary transport results into the trained artificial intelligence model, extracting the spatial features and temporal features of the preliminary transport results through the convolutional layer and long short-term memory network of the artificial intelligence model, and extracting the local features of the preliminary transport results through the attention mechanism, and outputting the prediction data of the transport process of microplastics based on the spatial features, temporal features, and the local features; Compare the predicted microplastic transport process data with the actual microplastic transport process data in the historical microplastic data, and iteratively optimize the key parameters of the plume model based on the comparison results until the accuracy of the predicted microplastic transport process data is greater than the preset accuracy, to obtain an integrated and optimized plume model and an artificial intelligence model; Input the real-time collected environmental data and microplastic property data into the integrated and optimized plume model and artificial intelligence model to obtain the microplastic transport simulation results.

[0006] Optionally, the method further includes: Collect the real results of microplastic transport in real time, compare the microplastic transport simulation results with the real results of microplastic transport, and calculate the error between the microplastic transport simulation results and the real results of microplastic transport; Based on the calculated error, adjust the parameters of the integrated and optimized plume model and artificial intelligence model again until the error between the microplastic transport simulation results and the real results of microplastic transport is less than the error threshold, to obtain an integrated and optimized plume model and artificial intelligence model again.

[0007] Optionally, the method further includes: Before inputting the microplastic property data and the environmental data into the plume model, determine the quartiles of the microplastic property data and the environmental data respectively through the interquartile range IQR, to obtain the first quartile and the third quartile of the microplastic property data, and the first quartile and the third quartile of the environmental data; Calculate the first difference between the third quartile and the first quartile of the microplastic property data, and the second difference between the third quartile and the first quartile of the environmental data respectively; Based on the first quartile of the microplastic property data and the first difference, determine the first lower bound data of the microplastic property data, and based on the first quartile of the environmental data and the second difference, determine the second lower bound data of the environmental data; Based on the third quartile of the microplastic property data and the first difference, determine the first upper bound data of the microplastic property data, and based on the third quartile of the environmental data and the second difference, determine the second upper bound data of the environmental data; Determine the microplastic property data less than the first lower bound data and the microplastic property data greater than the first upper bound data as abnormal microplastic property data, and determine the environmental data less than the second lower bound data and the environmental data greater than the second upper bound data as abnormal environmental data; Accordingly, inputting the microplastic property data and the environmental data into the plume model includes: The microplastic property data and environmental data after removing the abnormal microplastic property data and the abnormal environmental data are input into the plume model.

[0008] Optionally, the microplastic transport simulation results include the transport path and diffusion range of microplastics under different conditions, and the different conditions include various extreme weather conditions.

[0009] Optionally, the method further includes: The microplastic transport simulation results are displayed through a visualization platform, and the microplastic transport simulation results include the flux of microplastics into the sea, distribution map, diffusion path and sedimentation situation; the visualization platform includes support for dynamic charts and three-dimensional geographic information systems.

[0010] Optionally, the microplastic attribute data is obtained through laboratory analysis and on-site monitoring, and the microplastic attribute data includes particle size, morphology, density and concentration of the microplastics; The environmental data is obtained through meteorological stations, hydrological monitoring stations and online water quality monitoring equipment, and the environmental data includes river flow, flow rate, tide, wind speed and meteorological data.

[0011] In a second aspect, an embodiment of the present invention provides a microplastics transport simulation system based on a plume model and an artificial intelligence model, the system comprising: Data collection module, used to collect microplastic property data, environmental data and historical microplastic transport data in the estuary area; A plume model construction module is used to construct a plume model based on the hydrological conditions of the estuary area and the characteristics of microplastics, wherein the plume model takes into account the influence of environmental factors on the transport process of microplastics; A plume model simulation module is used to input the microplastic property data and the environmental data into the plume model, simulate the transport process of microplastics through numerical calculation methods, and obtain preliminary transport results; An artificial intelligence module training module, used to train an artificial intelligence model based on the environmental data and the historical microplastic transport data, so that the trained artificial intelligence model can predict the microplastic transport process; A plume model and artificial intelligence model integrated optimization module is used to input the preliminary transport results into the trained artificial intelligence model, extract the spatial and temporal features of the preliminary transport results through the convolutional layer and long short-term memory network of the artificial intelligence model, and extract the local features of the preliminary transport results through the attention mechanism, and output microplastic transport process prediction data based on the spatial features, temporal features and local features; The plume model and artificial intelligence model integrated optimization module is further used to compare the predicted data of the microplastic transport process with the actual data of the microplastic transport process in the microplastic historical data, and iteratively optimize the key parameters of the plume model based on the comparison results until the accuracy of the predicted data of the microplastic transport process is greater than a preset accuracy, thereby obtaining an integrated optimized plume model and artificial intelligence model; The microplastic transport simulation module is used to input the real-time collected environmental data and microplastic property data into the integrated and optimized plume model and artificial intelligence model to obtain the microplastic transport simulation results.

[0012] In a third aspect, an embodiment of the present invention provides an electronic device, including: at least one processor; a memory for storing the at least one processor-executable instruction; The at least one processor is configured to execute the instructions to implement the method described in the first aspect.

[0013] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, which, when instructions in the computer-readable storage medium are executed by a processor of an electronic device, enables the electronic device to execute the method described in the first aspect.

[0014] In a fifth aspect, an embodiment of the present invention provides a computer program product, including a computer program, which implements the method described in the first aspect when executed by a processor.

[0015] The technical solution provided by the embodiments of the present invention constructs a plume model based on the hydrological conditions and microplastic characteristics of the estuary area. This plume model takes into account the impact of environmental factors on the microplastic transport process. Microplastic attribute data and the environmental data are input into the constructed plume model to obtain preliminary transport results. However, due to the shortcomings of the plume model in handling complex nonlinear problems, the accuracy of the preliminary transport results is not high enough. Therefore, the preliminary transport results are further input into the trained artificial intelligence model, which is capable of predicting the microplastic transport process. By capturing the spatial, temporal, and local characteristics of the preliminary transport results, the trained artificial intelligence model can optimize the preliminary transport results and output predicted data for the microplastic transport process. Finally, based on the comparison results of the predicted microplastic transport process data with the actual microplastic transport process data in the microplastic historical data, the key parameters of the plume model are continuously optimized until a highly accurate, integrated and optimized plume model and artificial intelligence model are obtained. Thus, the integrated optimized plume model and artificial intelligence model can be used to more accurately simulate the transport of microplastics into the sea.

[0016] It can be seen that the present invention realizes the accurate transport simulation of microplastics in the estuary area by combining the plume numerical model with modern artificial intelligence technology, providing effective technical support for the control of microplastic pollution and ecological risk assessment. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 A flow chart of a method for simulating the transport of microplastics into the sea based on a plume model and an artificial intelligence model provided by an embodiment of the present invention; Figure 2 A flowchart of a specific implementation method for identifying outliers in microplastic attribute data and environmental data; Figure 3 A schematic diagram of a microplastics transport simulation system into the sea based on a plume model and an artificial intelligence model provided by an embodiment of the present invention; Figure 4 A schematic diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0018] The present invention will be described in detail below through examples.

[0019] Microplastics, as emerging pollutants, have a significant impact on marine ecosystems and biodiversity. Accurate simulation of microplastic transport characteristics and fluxes into the ocean is fundamental to assessing their impact, predicting pollution trends, and designing pollution prevention and control measures.

[0020] Currently, mathematical models can be used to simulate the transport of microplastics. However, because the behavior of microplastics in the environment is affected by multiple factors, such as tides and wind, current mathematical models have difficulty capturing these complex dynamic changes, resulting in low accuracy in microplastic transport simulation results and limited applicability.

[0021] To address the aforementioned technical issues of the existing technology, the present invention provides a method for simulating the transport of microplastics from estuaries to the sea, combining a plume model with an artificial intelligence model. The artificial intelligence model can autonomously discover and simulate the transport patterns of microplastics by learning from a large amount of historical data, thus overcoming the shortcomings of the plume model in handling complex nonlinear problems. By integrating and optimizing the plume model and the artificial intelligence model, the accuracy and adaptability of the simulation of microplastic transport into the sea can be improved, providing effective technical support for the control of microplastic pollution and ecological risk assessment.

[0022] The overall technical solution provided by the embodiment of the present invention is described below. The overall technical solution of the embodiment of the present invention mainly includes the following steps: S1: Construct a plume model. Specifically, a plume model is used to simulate the basic transport process of microplastics. This plume model mainly considers hydrodynamic factors such as water flow, wind, and tides, and calculates the movement characteristics of microplastic particles in the water body, such as diffusion, sedimentation, and floating. Through refined plume model calculations, preliminary transport results of microplastics in the estuary area are obtained.

[0023] S2: Introduction of an artificial intelligence model. Specifically, this invention introduces machine learning (e.g., deep learning and support vector machines, etc.) models. By learning a large amount of historical microplastic transport data, the simulation results of the plume model are optimized, that is, the preliminary transport results obtained in S1 are optimized. The artificial intelligence model can identify non-linear relationships that are difficult to capture by the plume model, such as the behavioral changes of microplastics under different hydrological conditions, thereby improving the accuracy of the transport simulation results.

[0024] S3: Data-driven model optimization. The artificial intelligence model uses the measured data and simulation results of historical microplastics to optimize the parameter settings in the plume model through means such as regression analysis and data mining. Especially in complex environments, the artificial intelligence model can better model the dynamic changes of microplastics and achieve adaptive adjustment of the plume model.

[0025] S4: Multi-factor joint simulation.

[0026] During the transport simulation process, considering various environmental factors including water flow, tides, wind, and meteorology, the artificial intelligence model can automatically analyze the transport laws of microplastics under different conditions and finally output relatively accurate microplastic transport simulation results, providing precise data support for subsequent pollution prevention and control.

[0027] It can be seen that this invention proposes a method for simulating the transport of estuarine microplastics into the sea based on the combination of a plume model and an artificial intelligence model. This method combines the plume numerical model with modern artificial intelligence technology to achieve precise transport simulation of microplastics in the estuary area, providing effective technical support for the control of microplastic pollution and ecological risk assessment.

[0028] After elaborating on the overall technical solution of the embodiments of this invention, the following will elaborate in detail on a method for simulating the transport of microplastics into the sea based on a plume model and an artificial intelligence model provided by the embodiments of this invention.

[0029] It should be noted that the method for simulating the transport of microplastics into the sea based on a plume model and an artificial intelligence model provided by the embodiments of this invention can be applied to a system for simulating the transport of microplastics into the sea based on a plume model and an artificial intelligence model. This system can include a data acquisition and processing module, a plume model construction module, an artificial intelligence optimization module, and a result display module.

[0030] In a first aspect, an embodiment of the present invention provides a method for simulating the transport of microplastics into the sea based on a plume model and an artificial intelligence model, as Figure 1 shown, which may include the following steps: S110, collect microplastic property data, environmental data, and microplastic historical transport data in the estuary area.

[0031] Specifically, the data collection and preprocessing module first needs to collect relevant data on microplastics and the environment in the estuary area. The data collection frequency can be adjusted according to factors such as different seasons, meteorological conditions, and extreme weather to ensure the representativeness and comprehensiveness of the microplastic property data and environmental data. The data collected by the data collection and preprocessing module includes the following parts: Microplastic property data: Conduct on-site sampling from the estuary area (this estuary area can be any estuary area, such as the Nandu River in Hainan Island) to obtain basic property data of microplastics such as particle size, morphology, density, and concentration. In practical applications, these microplastic property data are obtained through laboratory analysis and on-site monitoring.

[0032] Environmental data: May include river flow rate, velocity, tide, wind speed, meteorological data, and hydro-meteorology, etc. This environmental data is obtained through meteorological stations, hydrological monitoring stations, and online water quality monitoring equipment, etc., and is stored in a time series manner.

[0033] Historical transport data: Obtain the transport data of microplastics in the estuary area through historical experiments and research. These data are used to train the artificial intelligence model.

[0034] S120, construct a plume model based on the hydrological conditions and microplastic characteristics in the estuary area.

[0035] Among them, the plume model takes into account the influence of environmental factors on the microplastic transport process.

[0036] S130, input the microplastic property data and environmental data into the plume model, and simulate the transport process of microplastics through numerical calculation methods to obtain preliminary transport results.

[0037] The plume model is mainly used to describe the movement characteristics of microplastics in water bodies such as transport, diffusion, and sedimentation. Specifically, it may include the following steps: S131, plume model design. Design a plume model according to the hydrological conditions of the target estuary and the characteristics of microplastics. The basic formula of the plume model is based on the principles of fluid mechanics, taking into account the influence of factors such as water flow, wind, and tide.

[0038] S132, Input data setting. The inputs of the plume model include environmental data and microplastic data, which can specifically include the water flow velocity, wind force, tidal cycle, particle size, density, flow characteristics, etc. of the estuary area. These input data can be set through on-site sampling data and meteorological monitoring data.

[0039] As an implementation manner of the embodiment of the present invention, in order to identify outliers in the microplastic attribute data and environmental data, as Figure 2 shown, the method may further include the following steps: S130a, Before inputting the microplastic attribute data and environmental data into the plume model, respectively determine the quartiles of the microplastic attribute data and environmental data through the interquartile range IQR, obtain the first quartile and the third quartile of the microplastic attribute data, and the first quartile and the third quartile of the environmental data.

[0040] S130b, Calculate the first difference between the third quartile and the first quartile of the microplastic attribute data, and the second difference between the third quartile and the first quartile of the environmental data, respectively.

[0041] S130c, Based on the first quartile and the first difference of the microplastic attribute data, determine the first lower bound data of the microplastic attribute data, and based on the first quartile and the second difference of the environmental data, determine the second lower bound data of the environmental data.

[0042] S130d, Based on the third quartile and the first difference of the microplastic attribute data, determine the first upper bound data of the microplastic attribute data, and based on the third quartile and the second difference of the environmental data, determine the second upper bound data of the environmental data.

[0043] S130e, Determine the abnormal microplastic attribute data for the microplastic attribute data less than the first lower bound data and greater than the first upper bound data, and determine the abnormal environmental data for the environmental data less than the second lower bound data and greater than the second upper bound data.

[0044] Correspondingly, S130, inputting the microplastic attribute data and environmental data into the plume model may include the following steps: Input the microplastic attribute data and environmental data after removing the abnormal microplastic attribute data and abnormal environmental data into the plume model.

[0045] Specifically, the IQR scoring method is used to identify outliers in the microplastic property data and environmental data. For the microplastic property dataset and the environmental dataset, the IQR scoring method can be used to extract the outliers respectively. The IQR divides the dataset into four equal parts using the quartiles in the statistical distribution, with each part containing 25% of the data points. The quartiles include the first quartile and the third quartile, where the first quartile can be represented by Q1 and the third quartile can be represented by Q3. The IQR is the difference between the third quartile and the first quartile (i.e., Q3 - Q1), which can effectively reflect the distribution range of the middle 50% of the data in the dataset.

[0046] Next, determine the lower bound of the data, and its calculation formula is: Q1 - 1.5 × IQR.

[0047] Determine the upper bound of the data, and its calculation formula is: Q3 + 1.5 × IQR.

[0048] After determining the lower bound and the upper bound of the data, the data less than the lower bound can be determined as abnormal data, and the data greater than the upper bound can be determined as abnormal data, thus realizing the identification and deletion of the abnormal data in the microplastic property data and the environmental data. Finally, the microplastic property data and the environmental data after removing the abnormal data are input into the plume model, and the transport process of microplastics is simulated by numerical methods, and the obtained preliminary transport results are more accurate. Among them, the preliminary transport results include the movement paths such as the diffusion and sedimentation of microplastics under different hydrological conditions.

[0049] S140. Based on the environmental data and the historical transport data of microplastics, train an artificial intelligence model, and the trained artificial intelligence model can predict the microplastic transport process.

[0050] Specifically, regarding the preparation of the training data. Use the historical microplastic transport data (such as the microplastic input flux into the sea, distribution, concentration, etc.) and the historical environmental data (such as flow velocity, tide, wind speed, etc.) to construct the training dataset. Input these data into the artificial intelligence model for training, so that the model can learn the potential laws of microplastic transport.

[0051] Moreover, in order to ensure that all species characteristics in the ocean data are on the same scale, thus avoiding some characteristics being given too much weight in the calculation process, it is planned to normalize the historical microplastic transport data and the historical environmental data in the spatio-temporal dimension using a sliding window that changes over time. After each sequence is shifted and scaled and input, the number of sequences and the number of variables are obtained respectively. The number of sequences refers to the number of time series in the time series dataset. The number of variables refers to the number of different variables contained in each time series.

[0052] Use methods such as cross - validation to train an artificial intelligence model, adjust the hyperparameters in the model, and optimize the prediction ability of the artificial intelligence model. Through multiple rounds of training, continuously improve the artificial intelligence model to ensure its accurate prediction of the microplastic transport process.

[0053] S150, Input the preliminary transport result into the trained artificial intelligence model. Extract the spatial features and temporal features of the preliminary transport result through the convolutional layer and long short - term memory network of the artificial intelligence model, and extract the local features of the preliminary transport result through the attention mechanism. Output the microplastic transport process prediction data based on the spatial features, temporal features, and the local features.

[0054] S160, Compare the microplastic transport process prediction data with the real data of the microplastic transport process in the microplastic historical data. Iteratively optimize the key parameters of the plume model based on the comparison result until the accuracy of the microplastic transport process prediction data is greater than the preset accuracy, and obtain the integrated and optimized plume model and artificial intelligence model.

[0055] Specifically, after the plume model outputs the preliminary transport result, input the preliminary transport result into the trained artificial intelligence model. Extract the spatial features and temporal features of the preliminary transport result through the convolutional layer and long short - term memory network (LSTM) of the artificial intelligence model, and extract the local features of the preliminary transport result through the attention mechanism. Since the trained artificial intelligence model can predict the microplastic transport process, the trained artificial intelligence model generates microplastic transport process prediction data based on the spatial features, temporal features, and local features. By comparing the microplastic transport process prediction data with the real data of the microplastic transport process in the microplastic historical data, if the comparison result shows that there is a large difference between the microplastic transport process prediction data and the real data of the microplastic transport process in the microplastic historical data, it indicates that the integrated and optimized plume model and artificial intelligence model cannot accurately predict the microplastic transport process. Therefore, it is necessary to adjust the key parameters of the plume model, where the key parameters can include water flow velocity, sedimentation velocity of microplastics, diffusion coefficient, etc. The artificial intelligence model iteratively optimizes the plume model based on learning the non - linear law from the historical data. The preliminary transport result output by the plume model is more accurate. After inputting the more accurate preliminary transport result into the artificial intelligence model, the artificial intelligence model outputs more accurate microplastic transport process prediction data. By continuously iteratively optimizing the simulation result of the plume model until the accuracy of the output microplastic transport process prediction data is greater than the preset accuracy, the integrated and optimized plume model and artificial intelligence model are obtained. The preset accuracy can be determined according to the actual situation and is not specifically limited here.

[0056] S170, Input the real - time collected environmental data and microplastic property data into the integrated and optimized plume model and artificial intelligence model to obtain the microplastic transport simulation result.

[0057] Specifically, after obtaining the integrated and optimized plume model and artificial intelligence model, the actual simulation process is then carried out. According to the real-time collected environmental data and microplastic property data, the microplastic transport simulation is performed through the integrated and optimized plume model and artificial intelligence model to obtain the microplastic transport simulation results. During the simulation process, the transport paths and diffusion ranges of microplastics under different conditions (such as high water period, normal water period, low water period, and extreme weather such as typhoons) are automatically generated. Therefore, the microplastic transport simulation results include the transport paths and diffusion ranges of microplastics under different conditions, and the different conditions include high water period, normal water period, low water period, and various extreme weather such as typhoons.

[0058] Based on the above embodiments, the method may further include the following steps, namely step 11 and step 12: Step 11, collect the real results of microplastic transport in real time, compare the microplastic transport simulation results with the real results of microplastic transport, and calculate the error between the microplastic transport simulation results and the real results of microplastic transport.

[0059] Step 12, based on the calculated error, adjust the parameters of the integrated and optimized plume model and artificial intelligence model again until the error between the microplastic transport simulation results and the real results of microplastic transport is less than the error threshold, and obtain the plume model and artificial intelligence model that are integrated and optimized again.

[0060] Specifically, this process is a model verification and comparison process. By comparing the microplastic transport simulation results with the real results of microplastic transport obtained from actual sampling, the accuracy of collecting the real results of microplastic transport is verified. By calculating the error between the microplastic transport simulation results and the real results of microplastic transport, and further adjusting and optimizing the integrated and optimized plume model and artificial intelligence model according to the error, until the error between the microplastic transport simulation results and the real results of microplastic transport is less than the error threshold, the plume model and artificial intelligence model that are integrated and optimized again are obtained, and the accuracy of the microplastic transport simulation results output by the plume model and artificial intelligence model that are integrated and optimized again is higher.

[0061] The technical solution provided by the embodiments of the present invention constructs a plume model based on the hydrological conditions and microplastic characteristics of the estuary area. This plume model takes into account the impact of environmental factors on the microplastic transport process. Microplastic attribute data and the environmental data are input into the constructed plume model to obtain preliminary transport results. However, due to the shortcomings of the plume model in handling complex nonlinear problems, the accuracy of the preliminary transport results is not high enough. Therefore, the preliminary transport results are further input into the trained artificial intelligence model, which is capable of predicting the microplastic transport process. By capturing the spatial, temporal, and local characteristics of the preliminary transport results, the trained artificial intelligence model can optimize the preliminary transport results and output predicted data for the microplastic transport process. Finally, based on the comparison results of the predicted microplastic transport process data with the actual microplastic transport process data in the microplastic historical data, the key parameters of the plume model are continuously optimized until a highly accurate, integrated and optimized plume model and artificial intelligence model are obtained. Thus, the integrated optimized plume model and artificial intelligence model can be used to more accurately simulate the transport of microplastics into the sea.

[0062] It can be seen that the present invention realizes the accurate transport simulation of microplastics in the estuary area by combining the plume numerical model with modern artificial intelligence technology, providing effective technical support for the control of microplastic pollution and ecological risk assessment.

[0063] Based on the above embodiment, as an implementation of the embodiment of the present invention, the method may further include the following steps: The microplastic transport simulation results are displayed through a visualization platform.

[0064] Among them, the simulation results of microplastic transport include the flux, distribution map, diffusion path and sedimentation of microplastics into the sea; the visualization platform includes support for dynamic charts and three-dimensional geographic information systems.

[0065] By visually displaying the microplastic transport simulation results, users can conduct data analysis and make decisions more easily.

[0066] In the second aspect, the embodiment of the present invention provides a microplastics transport simulation system based on plume model and artificial intelligence model, such as Figure 3 As shown, the system includes: Data collection module 310, for collecting microplastic property data, environmental data, and historical microplastic transport data in the estuary area; A plume model construction module 320 is used to construct a plume model based on the hydrological conditions of the estuary area and the characteristics of microplastics, wherein the plume model takes into account the impact of environmental factors on the transport process of microplastics; The plume model simulation module 330 is configured to input the microplastic property data and the environmental data into a plume model, simulate the transport process of microplastics by means of numerical calculation methods, and obtain preliminary transport results; The artificial intelligence module training module 340 is configured to train an artificial intelligence model based on the environmental data and the historical microplastic transport data. The trained artificial intelligence model can predict the microplastic transport process; The plume model and artificial intelligence model integration and optimization module 350 is configured to input the preliminary transport results into the trained artificial intelligence model, extract the spatial features and temporal features of the preliminary transport results through the convolutional layer and long short-term memory network of the artificial intelligence model, and extract the local features of the preliminary transport results through an attention mechanism, and output microplastic transport process prediction data based on the spatial features, temporal features, and the local features; The plume model and artificial intelligence model integration and optimization module 350 is further configured to compare the microplastic transport process prediction data with the real data of the microplastic transport process in the microplastic historical data, and iteratively optimize the key parameters of the plume model based on the comparison results until the accuracy of the microplastic transport process prediction data is greater than a preset accuracy, so as to obtain an integrated and optimized plume model and artificial intelligence model; The microplastic transport simulation module 360 is configured to input the real-time collected environmental data and microplastic property data into the integrated and optimized plume model and artificial intelligence model to obtain microplastic transport simulation results.

[0067] In a third aspect, an embodiment of the present invention provides an electronic device, as Figure 4 shown, including: At least one processor 401; A memory 402 for storing instructions executable by the at least one processor; Wherein, the at least one processor is configured to execute the instructions to implement the method described in the first aspect.

[0068] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium. When the instructions in the computer-readable storage medium are executed by a processor of an electronic device, the electronic device can execute the method described in the first aspect.

[0069] In a fifth aspect, an embodiment of the present invention provides a computer program product, including a computer program, where the computer program implements the method described in the first aspect when executed by a processor.

[0070] Although the embodiments of the present invention have been shown and described above, it can be understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those of ordinary skill in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present invention without departing from the principles and spirit of the present invention.

Claims

1. A simulation method for the transport of microplastics into the sea based on the plume model and the artificial intelligence model, characterized in that, The method includes: Collecting microplastic property data, environmental data, and microplastic historical transport data in the estuary area; Constructing a plume model based on the hydrological conditions and microplastic characteristics in the estuary area, where the plume model takes into account the influence of environmental factors on the microplastic transport process; Inputting the microplastic property data and the environmental data into the plume model, and simulating the microplastic transport process through numerical calculation methods to obtain a preliminary transport result; Training an artificial intelligence model based on the environmental data and the microplastic historical transport data, and the trained artificial intelligence model can predict the microplastic transport process; Inputting the preliminary transport result into the trained artificial intelligence model, extracting the spatial features and temporal features of the preliminary transport result through the convolutional layer and long short-term memory network of the artificial intelligence model, and extracting the local features of the preliminary transport result through the attention mechanism, and outputting microplastic transport process prediction data based on the spatial features, temporal features, and the local features; Comparing the microplastic transport process prediction data with the real data of the microplastic transport process in the microplastic historical data, and iteratively optimizing the key parameters of the plume model based on the comparison result until the accuracy of the microplastic transport process prediction data is greater than the preset accuracy, to obtain an integrated and optimized plume model and artificial intelligence model; Inputting the real-time collected environmental data and microplastic property data into the integrated and optimized plume model and artificial intelligence model to obtain a microplastic transport simulation result.

2. The method according to claim 1, characterized in that, The method further includes: Collecting the real result of microplastic transport in real time, comparing the microplastic transport simulation result with the real result of microplastic transport, and calculating the error between the microplastic transport simulation result and the real result of microplastic transport; Based on the calculated error, adjusting the parameters of the integrated and optimized plume model and artificial intelligence model again until the error between the microplastic transport simulation result and the real result of microplastic transport is less than the error threshold, to obtain a plume model and artificial intelligence model that are integrated and optimized again.

3. The method according to claim 1, wherein The method further includes: Before inputting the microplastic property data and the environmental data into the plume model, respectively determining the quartiles of the microplastic property data and the environmental data through the interquartile range IQR, to obtain the first quartile and the third quartile of the microplastic property data, and the first quartile and the third quartile of the environmental data; Calculating the first difference between the third quartile and the first quartile of the microplastic property data, and the second difference between the third quartile and the first quartile of the environmental data respectively; Based on the first quartile of the microplastic property data and the first difference, determining the first lower bound data of the microplastic property data, and based on the first quartile of the environmental data and the second difference, determining the second lower bound data of the environmental data; Determine a first upper limit data of the microplastic attribute data based on the third quartile of the microplastic attribute data and the first difference, and determine a second upper limit data of the environmental data based on the third quartile of the environmental data and the second difference; Determine the microplastic attribute data that is less than the first lower limit data and the microplastic attribute data that is greater than the first upper limit data as abnormal microplastic attribute data, and determine the environmental data that is less than the second lower limit data and the environmental data that is greater than the second upper limit data as abnormal environmental data; Accordingly, inputting the microplastic property data and the environmental data into the plume model includes: The microplastic property data and environmental data after removing the abnormal microplastic property data and the abnormal environmental data are input into the plume model.

4. The method according to any one of claims 1 to 3, characterized in that The microplastic transport simulation results include the transport path and diffusion range of microplastics under different conditions, and the different conditions include various extreme weather conditions.

5. The method according to any one of claims 1 to 3, characterized in that The method further comprises: The microplastic transport simulation results are displayed through a visualization platform, and the microplastic transport simulation results include the flux of microplastics into the sea, distribution map, diffusion path and sedimentation situation; the visualization platform includes support for dynamic charts and three-dimensional geographic information systems.

6. The method according to any one of claims 1 to 3, characterized in that The microplastic attribute data is obtained through laboratory analysis and on-site monitoring, and the microplastic attribute data includes the particle size, morphology, density and concentration of the microplastics; The environmental data is obtained through meteorological stations, hydrological monitoring stations and online water quality monitoring equipment, and the environmental data includes river flow, flow rate, tide, wind speed and meteorological data.

7. A microplastic ocean transport simulation system based on a plume model and an artificial intelligence model, characterized in that, The system comprises: Data collection module, used to collect microplastic property data, environmental data and historical microplastic transport data in the estuary area; A plume model construction module is used to construct a plume model based on the hydrological conditions of the estuary area and the characteristics of microplastics, wherein the plume model takes into account the influence of environmental factors on the transport process of microplastics; A plume model simulation module is used to input the microplastic property data and the environmental data into the plume model, simulate the transport process of microplastics through numerical calculation methods, and obtain preliminary transport results; An artificial intelligence module training module, used to train an artificial intelligence model based on the environmental data and the historical microplastic transport data, so that the trained artificial intelligence model can predict the microplastic transport process; A plume model and artificial intelligence model integrated optimization module is used to input the preliminary transport results into the trained artificial intelligence model, extract the spatial and temporal features of the preliminary transport results through the convolutional layer and long short-term memory network of the artificial intelligence model, and extract the local features of the preliminary transport results through the attention mechanism, and output microplastic transport process prediction data based on the spatial features, temporal features and local features; The plume model and artificial intelligence model integrated optimization module is also used to compare the predicted microplastic transport process data with the actual microplastic transport process data in the microplastic historical data, and iteratively optimize the key parameters of the plume model based on the comparison results until the accuracy of the predicted microplastic transport process data is greater than the preset accuracy, so as to obtain the integrated and optimized plume model and artificial intelligence model; The microplastic transport simulation module is used to input the real-time collected environmental data and microplastic attribute data into the integrated and optimized plume model and artificial intelligence model to obtain the microplastic transport simulation results.

8. An electronic device, characterized in that, Comprising: At least one processor; A memory for storing executable instructions of the at least one processor; Wherein, the at least one processor is configured to execute the instructions to implement the method according to any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, When the instructions in the computer-readable storage medium are executed by the processor of the electronic device, the electronic device can execute the method according to any one of claims 1-6.

10. A computer program product, characterized in that, Comprising a computer program, which implements the method according to any one of claims 1-6 when executed by a processor.

Images