Fluorinated industrial park pfas prediction system based on multi-medium dynamic allocation and collaborative verification method

Abstract

The application discloses a fluorine chemical industry park PFAS prediction system and a collaborative verification method based on multi-medium dynamic distribution, relates to the technical field of environmental monitoring and pollution control, and comprises a fluorine chemical industry data acquisition module, a fluorine chemical industry feature extraction module, a cross-medium correlation distribution module, a PFAS concentration risk prediction module and a PFAS visual display module; the fluorine chemical industry data acquisition module acquires and pre-processes chemical oxygen demand data and fluorine ion concentration data in water bodies at discharge outlets of a fluorine chemical industry park, and detects the PFAS concentration in water bodies, sediments and aquatic organisms by sampling. The application effectively improves the monitoring and early warning capability of PFAS pollution in the fluorine chemical industry park by constructing the PFAS prediction system and the collaborative verification method based on multi-medium dynamic distribution, realizes accurate prediction of the PFAS concentration and rapid identification of pollution sources, reduces the detection cost, improves the real-time performance and the dynamic early warning level, provides a scientific basis for environmental pollution control, and promotes the development of the environment-friendly fluorine chemical industry.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of environmental monitoring and pollution control technology, in particular to a fluorine chemical industry park PFAS prediction system and collaborative verification method based on multi-medium dynamic distribution. BACKGROUND

[0002] PFAS, perfluoroalkyl and polyfluoroalkyl substances, are a series of non-natural synthetic organic compounds, which are mainly composed of carbon atoms and fluorine atoms, and have extremely high thermal stability and chemical stability, so they are widely used in industrial production and consumer fields such as textiles, surfactants, food packaging, non-stick coatings, and fire extinguishing foam. PFAS is not easily biodegradable and can accumulate in the environment and organisms. Contact with certain types of PFAS may have serious effects on human health, such as endocrine function disorders, cancer, thyroid disease, and liver and kidney damage. In addition, PFAS includes a variety of specific compounds such as perfluorooctane sulfonic acid, perfluorooctanoic acid, and perfluorohexyl sulfonic acid. These compounds have high detection rates and concentration levels in the environment and have different toxic effects on organisms. Due to the long-term exposure and potential health risks of PFAS, it has attracted widespread attention worldwide, and many countries have begun to strictly regulate and limit PFAS. Therefore, understanding the properties, applications, and health effects of PFAS is of great significance for protecting human health and environmental safety.

[0003] In order to solve the problems of high detection cost, multi-medium dynamic distribution prediction not yet realized, and insufficient real-time and dynamic early warning capability of PFAS, the existing technology mainly relies on traditional laboratory detection methods, which monitor the distribution of PFAS in different environmental media through regular sampling and analysis. However, this method is not only time-consuming and labor-intensive, but also costly, making it difficult to achieve real-time monitoring and dynamic early warning of PFAS pollution conditions. In addition, due to the complex behavior of PFAS in the environment, involving dynamic distribution between multiple media, traditional prediction methods often fail to accurately predict its migration and transformation rules, thereby failing to effectively guide PFAS pollution control and risk management. In order to overcome these limitations, a fluorine chemical industry park PFAS prediction system and collaborative verification method based on multi-medium dynamic distribution are proposed. SUMMARY

[0004] The present application aims to provide a fluorine chemical industry park PFAS prediction system and collaborative verification method based on multi-medium dynamic distribution to solve the problems raised in the background art.

[0005] To solve the above technical problems, the technical scheme adopted by the present application is: in the first aspect, a PFAS prediction system for a fluorine chemical industrial park based on multi-medium dynamic allocation, comprising a fluorine chemical industrial data acquisition module, a fluorine chemical industrial feature extraction module, a cross-medium correlation allocation module, a PFAS concentration risk prediction module, and a PFAS visualization display module;

[0006] The fluorine chemical industrial data acquisition module acquires and pre-processes chemical oxygen demand data and fluorine ion concentration data in water bodies at discharge outlets of the fluorine chemical industrial park, and samples and detects PFAS concentrations in water bodies, sediments, and aquatic organisms;

[0007] The fluorine chemical industrial feature extraction module extracts fluorine chemical industrial features from the pre-processed chemical oxygen demand data and fluorine ion concentration data;

[0008] The cross-medium correlation allocation module, in combination with the extracted fluorine chemical industrial features, constructs a cross-medium correlation model and a multi-medium dynamic allocation model to predict concentrations of PFAS in water bodies, sediments, and aquatic organisms;

[0009] The PFAS concentration risk prediction module, based on the predicted values of the concentrations of PFAS in water bodies, sediments, and aquatic organisms, analyzes whether the predicted values of the concentrations of PFAS in each medium deviate from preset reference values;

[0010] The PFAS visualization display module visually displays spatial distribution of the concentrations of PFAS in water bodies, sediments, and aquatic organisms, and compares historical prediction data with actual detection data.

[0011] In the second aspect, a PFAS prediction system for a fluorine chemical industrial park based on multi-medium dynamic allocation and a collaborative verification method, comprising the following steps:

[0012] S1, acquire and pre-process chemical oxygen demand data and fluorine ion concentration data in water bodies at discharge outlets of the fluorine chemical industrial park, and sample and detect PFAS concentrations in water bodies, sediments, and aquatic organisms;

[0013] S2, extract fluorine chemical industrial features based on the pre-processed chemical oxygen demand data and fluorine ion concentration data;

[0014] S3, in combination with the extracted fluorine chemical industrial features, construct a cross-medium correlation model and a multi-medium dynamic allocation model to respectively predict concentrations of PFAS in water bodies, sediments, and aquatic organisms;

[0015] S4, based on historical data of the concentrations of PFAS in water bodies, sediments, and aquatic organisms, preset reference values of the concentrations of PFAS in each medium, and compare the preset reference values with the predicted values of the concentrations of PFAS in water bodies, sediments, and aquatic organisms, analyze whether there is a deviation from the reference values;

[0016] S5, the concentration of PFAS in water, sediment and aquatic organisms, spatial distribution, historical prediction and actual detection data comparison visualization display.

[0017] Further improvement of the technical scheme of the application lies in that in S1, the collection and preprocessing process of the chemical oxygen demand data and the fluoride ion concentration data in the water body at the discharge outlet of the fluorine chemical industry park includes:

[0018] The ultraviolet spectroscopy online COD analyzer and the fluoride ion selective electrode sensor are arranged in the monitoring well of the main channel of the discharge outlet of the fluorine chemical industry park. The ultraviolet light source emits a 254nm wavelength light beam to penetrate the water sample, detect the attenuation intensity of the light beam, collect the chemical oxygen demand data in the water body at the discharge outlet of the fluorine chemical industry park according to the Lambert-Beer law, and the electrode surface fluoride ion selective membrane of the fluoride ion selective electrode sensor has a potential response with the fluoride ion in the water sample. The fluoride ion concentration data is obtained through the Nernst equation;

[0019] For the chemical oxygen demand data and the fluoride ion concentration data collected every hour, the 3σ criterion is used to identify abnormal values. If the data point is outside the range of mean value ± 3 times standard deviation, it is marked as invalid data and removed. For data missing for no more than 2 hours continuously, the linear interpolation method of adjacent time points is used to fill in the data. The collected chemical oxygen demand data and fluoride ion concentration data are scaled to the [0, 1] interval through Min-Max standardization.

[0020] Further improvement of the technical scheme of the application lies in that in S1, the process of sampling and detecting the concentration of PFAS in water, sediment and aquatic organisms includes:

[0021] The automatic water quality sampler, gravity type columnar sediment sampler and trawl type biological sampler are arranged downstream of the discharge outlet of the fluorine chemical industry park. 1 liter of water, 500 grams of sediment and 50-100 grams of aquatic organisms are collected every month. Liquid chromatography-mass spectrometry is used to detect PFAS, equipped with a solid-phase extraction column and an ion source. Before detection, the water sample is fixed by nitric acid, the sediment is ultrasonically extracted by acetonitrile, and the biological sample is extracted by accelerated solvent.

[0022] In mass spectrometry, after the water sample, sediment sample and aquatic organism sample are separated by liquid chromatography, they enter the electrospray ion source in positive ion mode to generate PFAS parent ions. A multiple reaction monitoring mode is used to set a specific mass-to-charge ratio. The PFAS parent ion intensity is recorded by the mass spectrometer, and the concentration of PFAS in water, sediment and aquatic organisms is calibrated based on the internal standard method.

[0023] Further improvement of the technical scheme of the application lies in that in S2, the process of extracting fluorine chemical characteristics based on the preprocessed chemical oxygen demand data and fluoride ion concentration data includes:

[0024] The fluorine chemical industry features include a chemical oxygen demand dynamic accumulation index and a fluorine ion concentration fluctuation rate;

[0025] Based on the pretreated chemical oxygen demand data per hour, the arithmetic mean and the standard deviation of the chemical oxygen demand data in the window are counted in a 24-hour sliding window period, the mean value is added to 2 times the standard deviation to obtain the chemical oxygen demand dynamic accumulation index;

[0026] The mean value and the standard deviation of the pretreated fluorine ion concentration data are counted per month, and the fluorine ion concentration fluctuation rate is obtained through the coefficient of variation formula.

[0027] The further improvement of the technical scheme of the present application is that in S3, the process of constructing a cross-medium correlation model to predict the concentration of PFAS in the water body comprises:

[0028] A cross-medium correlation model is constructed based on a long short-term memory network architecture, the extracted fluorine chemical industry features are input into the input layer of the cross-medium correlation model, and seasonal encoding and real-time flow data are introduced;

[0029] The hidden layer neurons of the cross-medium correlation model capture the dynamic correlation between the fluorine chemical industry features and the concentration of PFAS in the water body, output the hidden state, the seasonal encoding and the real-time flow data are fused through a fully connected layer and the hidden state, and the predicted value of the concentration of PFAS in the water body is output, and the predicted value of the concentration of PFAS in the water body is constrained according to the water-sediment-biota multi-medium distribution rule.

[0030] The further improvement of the technical scheme of the present application is that in S3, the process of constructing a multi-medium dynamic distribution model to predict the concentration of PFAS in the sediment and the water organism comprises:

[0031] The biological enrichment coefficient BCF is determined by fitting the ratio of the PFAS concentrations of the organism and the sediment in the historical data, the predicted value of the concentration of PFAS in the water body output by the cross-medium correlation model is combined, a multi-medium dynamic distribution model is constructed based on the law of conservation of mass, and the concentrations of PFAS in the sediment and the water organism are calculated.

[0032] The further improvement of the technical scheme of the present application is that S4 specifically comprises:

[0033] Based on the contribution of the water body, the sediment and the water organism to the ecological risk in the historical data, the historical data of the same time length are analyzed, the average values of the PFAS concentrations of the water body, the sediment and the water organism are combined, and the reference values of the PFAS concentrations of the water body, the sediment and the water organism are determined;

[0034] The predicted value of the PFAS concentration of each medium is compared with the corresponding reference value, the deviation rate of the PFAS concentration of each medium is analyzed, whether the predicted value of the PFAS concentration of each medium is abnormal is judged, and then independent early warning of each medium is triggered.

[0035] Further improvement of the technical scheme of the application is that: in S5, the process of visualizing the comparison between the spatial distribution of the PFAS concentration and the predicted value and the actual detection value comprises:

[0036] Based on the coordinates of the monitoring points in the fluorine chemical industrial park and the predicted values of the PFAS concentrations of the corresponding water bodies, sediments and aquatic organisms, the inverse distance weighted method is used to introduce a water dynamic model to simulate the diffusion path of the pollutants, and the concentration field output by the water dynamic model is combined with the inverse distance weighted interpolation to interpolate the discrete point concentration data into continuous spatial distribution, and the interpolated concentration is mapped into color intensity according to the gradient, red represents high concentration, blue represents medium concentration, and green represents low concentration, to generate a park PFAS concentration spatial distribution heat map.

[0037] The historical prediction and actual detection data comparison is displayed using a line chart, the actual detection value, the predicted value and the actual value of the water bodies, sediments and aquatic organisms are aligned on a monthly time axis, the predicted values of each month are connected by a dashed line, the horizontal axis is time, and the vertical axis is concentration, the sampling detection values are connected by a solid line, and the predicted curve is displayed in superposition.

[0038] Due to the adoption of the above technical scheme, the technical progress achieved by the application relative to the prior art is:

[0039] 1. The application provides a fluorine chemical industrial park PFAS prediction system and a collaborative verification method based on multi-medium dynamic allocation, realizes the concentration prediction of PFAS in various environmental media, significantly improves the accuracy and real-time performance of the prediction, and provides a scientific basis for park environmental management.

[0040] 2. The application provides a fluorine chemical industrial park PFAS prediction system and a collaborative verification method based on multi-medium dynamic allocation, which visualizes the spatial distribution of the PFAS concentration, the prediction and the actual detection data comparison, provides intuitive data support for park environmental pollution control and management, and helps to develop more scientific and reasonable pollution control strategies. DETAILED DESCRIPTION

[0041] In order to more clearly illustrate the technical scheme in the embodiments of the present application or the prior art, the drawings needed in the embodiments will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments described in the present application, and other drawings can also be obtained by those skilled in the art based on these drawings.

[0042] Figure 1 The flowchart of the present application. Detailed Implementation

[0043] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0044] Examples, such as Figure 1 As shown, the present invention provides a PFAS prediction system for fluorochemical industrial parks based on multi-media dynamic allocation, including a fluorochemical data acquisition module, a fluorochemical feature extraction module, a cross-media association allocation module, a PFAS concentration risk prediction module, and a PFAS visualization module.

[0045] The fluorochemical data acquisition module collects and preprocesses chemical oxygen demand (COD) and fluoride ion concentration data in the water bodies at the discharge outlets of fluorochemical industrial parks, and also samples and tests the PFAS concentration in water bodies, sediments, and aquatic organisms.

[0046] The fluorochemical feature extraction module extracts fluorochemical features from preprocessed chemical oxygen demand (COD) and fluoride ion concentration (FOR) data.

[0047] The cross-media correlation and allocation module, combined with the extracted fluorochemical characteristics, constructs a cross-media correlation model and a multi-media dynamic allocation model to predict the concentration of PFAS in water bodies, sediments and aquatic organisms;

[0048] The PFAS concentration risk prediction module analyzes whether the predicted PFAS concentration in each medium deviates from the preset reference value based on the predicted concentration values ​​of PFAS in water, sediment and aquatic organisms.

[0049] The PFAS visualization module visualizes the spatial distribution of PFAS concentrations in water bodies, sediments, and aquatic organisms, and compares historical predictions with actual detection data.

[0050] Example 2, as Figure 1 As shown, based on Example 1, the present invention also provides a technical solution: a PFAS prediction system and collaborative verification method for fluorochemical industrial parks based on multi-media dynamic allocation, comprising the following steps:

[0051] S1, collect and pretreat the chemical oxygen demand data and the fluorine ion concentration data in the water body of the fluorine chemical industry park discharge port, detect the PFAS concentration in the water body, sediment and aquatic organism by sampling, deploy the ultraviolet spectrum method online COD analyzer and the fluorine ion selective electrode sensor in the monitoring well of the main channel of the fluorine chemical industry park discharge port, the ultraviolet light source emits 254nm wavelength light beam to penetrate the water sample, detect the intensity of the detection beam, according to the Lambert-Beer law, collect the chemical oxygen demand data in the water body of the fluorine chemical industry park discharge port, the fluorine ion selective membrane on the electrode surface of the fluorine ion selective electrode sensor has potential response with the fluorine ion in the water sample, obtain the fluorine ion concentration data through the Nernst equation, wherein the ultraviolet spectrum method online COD analyzer is 10 meters, 50 meters, 100 meters and 200 meters away from the downstream of the discharge port, and the automatic water quality sampler is deployed, the probe of which is vertically inserted into the water surface below 0.5 meters, and the data is automatically collected once an hour, the measurement range is 0-500ng / g, the fluorine ion selective electrode sensor is automatically calibrated once every 24 hours by standard sodium fluoride solution, the chemical oxygen demand data and the fluorine ion concentration data collected every hour are identified by 3σ criterion, if the data point is out of the range of mean value ± 3 times standard deviation, it is marked as invalid data and eliminated, for the data missing for no more than 2 hours, the linear interpolation method of adjacent time point data is used to fill in, the collected chemical oxygen demand data and fluorine ion concentration data are scaled to the interval [0, 1] through Min-Max standardization, the automatic water quality sampler, gravity type columnar sediment sampler and trawl type biological sampler are deployed downstream of the fluorine chemical industry park discharge port, 1 liter of water body, 500 grams of sediment and 50-100 grams of individual aquatic organisms are collected every month, liquid chromatography-mass spectrometry is used to detect PFAS, solid phase extraction column and ion source are equipped, before detection, the water sample is fixed by nitric acid, the sediment is ultrasonic extracted by acetonitrile and the biological sample is accelerated solvent extracted, in the mass spectrometry detection, the water sample, sediment sample and aquatic organism sample are separated by liquid chromatography and then enter the electrospray ion source, PFAS parent ions are generated in the positive ion mode, the multiple reaction monitoring mode is adopted, the specific mass-to-charge ratio is set, the PFAS parent ion intensity is recorded by the mass spectrometer, and the PFAS concentration in the water body, sediment and aquatic organism is calibrated based on the internal standard method;

[0052] S2, extract fluorine chemical characteristics based on the pretreated chemical oxygen demand data and fluorine ion concentration data, the fluorine chemical characteristics include chemical oxygen demand dynamic accumulation index and fluorine ion concentration fluctuation rate, based on the pretreated chemical oxygen demand data, take 24 hours as the sliding window period, calculate the arithmetic mean and standard deviation of the chemical oxygen demand data in the window, add the mean value to 2 times the standard deviation to obtain the chemical oxygen demand dynamic accumulation index, for the pretreated fluorine ion concentration data, calculate the mean value and standard deviation every month, and obtain the fluorine ion concentration fluctuation rate through the coefficient of variation formula;

[0053] S3, combined with the extracted fluorine chemical industry characteristics, a cross-media correlation model and a multi-medium dynamic allocation model are constructed to predict the concentration of PFAS in water, sediment and aquatic organisms. The cross-media correlation model is constructed based on the long short-term memory network architecture. The extracted fluorine chemical industry characteristics are input into the input layer of the cross-media correlation model. Seasonal encoding and real-time flow data are introduced. The hidden layer neurons of the cross-media correlation model capture the dynamic correlation between the fluorine chemical industry characteristics and the concentration of PFAS in water. The output hidden state is fused with the seasonal encoding and real-time flow data through a fully connected layer to output the predicted value of the concentration of PFAS in water. According to the water-sediment-organism multi-medium distribution rule, the predicted value of the concentration of PFAS in water is constrained. The biological enrichment coefficient is determined by fitting the ratio of the concentration of PFAS in organisms and sediment in the historical data. Based on the mass conservation law, a multi-medium dynamic allocation model is constructed to calculate the concentration of PFAS in sediment and aquatic organisms.

[0054] S4, based on the concentration of PFAS in water, sediment and aquatic organisms in the historical data, the reference value of the concentration of PFAS in each medium is preset, and compared with the predicted value of the concentration of PFAS in water, sediment and aquatic organisms. Whether there is deviation from the reference value is analyzed. Based on the contribution of water, sediment and aquatic organisms to ecological risk in the historical data, the historical data of the same time length is analyzed. The reference value of the concentration of PFAS in water, sediment and aquatic organisms is determined by combining the average value of the concentration of PFAS in water, sediment and aquatic organisms. The predicted value of the concentration of PFAS in each medium is compared with the corresponding reference value to analyze the deviation rate of the concentration of PFAS in each medium. Whether the predicted value of the concentration of PFAS in each medium is abnormal is judged, and then the independent warning of each medium is triggered.

[0055] S5, the spatial distribution of the concentration of PFAS in water, sediment and aquatic organisms, the comparison between the historical prediction and the actual detection data is visualized. Based on the coordinates of the monitoring points in the fluorine chemical industry park and the predicted values of the concentration of PFAS in water, sediment and aquatic organisms, the inverse distance weighted method is used to introduce the water dynamic model to simulate the diffusion path of pollutants. The concentration field output by the water dynamic model is combined with the inverse distance weighted interpolation to interpolate the discrete point concentration data into continuous spatial distribution. The interpolated concentration is mapped to color intensity according to gradient. Red represents high concentration, blue represents medium concentration, and green represents low concentration. The PFAS concentration spatial distribution heat map of the park is generated. The comparison between the historical prediction and the actual detection data is displayed using a line chart. The integrated risk index and the actual detection value are aligned on the monthly time axis. The predicted values of water, sediment and aquatic organisms are compared with the actual values. The predicted values of each month are connected by a dashed line. The horizontal axis represents time, and the vertical axis represents concentration. The sampling detection values are connected by a solid line, and the predicted curve is displayed in superposition.

[0056] The above description is only specific embodiments of the present application, but the protection scope of the present application is not limited thereto, any person skilled in the art can easily think of changes or replacements within the technical scope disclosed by the present application, which should be covered in the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A PFAS prediction system for fluorochemical industrial parks based on dynamic multi-media distribution, characterized in that: It includes a fluorochemical data acquisition module, a fluorochemical feature extraction module, a cross-media correlation and allocation module, a PFAS concentration risk prediction module, and a PFAS visualization module; The fluorochemical data acquisition module collects and preprocesses chemical oxygen demand (COD) and fluoride ion concentration data in the water body at the discharge outlet of the fluorochemical industrial park, and samples and detects the PFAS concentration in the water body, sediment and aquatic organisms. The fluorochemical feature extraction module extracts fluorochemical features from the pre-processed chemical oxygen demand data and fluoride ion concentration data. The cross-media association allocation module, combined with the extracted fluorochemical characteristics, constructs a cross-media association model and a multi-media dynamic allocation model to predict the concentration of PFAS in water, sediment, and aquatic organisms. Specifically, the cross-media association model is constructed based on a long short-term memory network architecture. The bioaccumulation coefficient is determined by fitting the ratio of PFAS concentration in organisms to sediments in historical data. Combined with the predicted PFAS concentration in water output by the cross-media association model, a multi-media dynamic allocation model is constructed based on the law of conservation of mass to calculate the PFAS concentration in sediments and aquatic organisms. The PFAS concentration risk prediction module analyzes whether the predicted PFAS concentration in each medium deviates from the preset reference value based on the predicted concentration values ​​of PFAS in water, sediment and aquatic organisms. The PFAS visualization module visualizes the spatial distribution of PFAS concentration in water bodies, sediments, and aquatic organisms, and compares historical predictions with actual detection data.

2. A method for collaborative verification of PFAS in fluorochemical industrial parks based on dynamic multi-media allocation, implemented based on the PFAS prediction system for fluorochemical industrial parks based on dynamic multi-media allocation as described in claim 1, characterized in that, It consists of the following steps: S1. Collect and preprocess chemical oxygen demand (COD) and fluoride ion concentration (FOS) data in the water body at the discharge outlet of the fluorochemical industrial park, and sample and test the PFAS concentration in the water body, sediment and aquatic organisms. S2. Extract fluorochemical characteristics based on preprocessed chemical oxygen demand (COD) and fluoride ion concentration (FOR) data; S3. Based on the extracted fluorine chemical characteristics, construct cross-media correlation models and multi-media dynamic allocation models to predict the concentration of PFAS in water bodies, sediments and aquatic organisms, respectively. S4. Based on historical data, the concentration values ​​of PFAS in water, sediment and aquatic organisms are used to preset reference values ​​for the concentration of PFAS in each medium, and these values ​​are compared with the predicted concentration values ​​of PFAS in water, sediment and aquatic organisms to analyze whether there are any deviations from the reference values. S5. Visualize the spatial distribution of PFAS concentrations in water bodies, sediments, and aquatic organisms, and compare historical predictions with actual detection data.

3. The method for collaborative verification of PFAS in fluorochemical industrial parks based on dynamic multi-media allocation according to claim 2, characterized in that: In S1, the process of collecting and preprocessing chemical oxygen demand (COD) and fluoride ion concentration (FOR) data in the water body at the discharge outlet of the fluorochemical industrial park includes: An online COD analyzer using ultraviolet spectroscopy and a fluoride ion selective electrode sensor are deployed side-by-side in the monitoring wells of the main channel of the fluorine chemical industrial park's discharge outlet. The ultraviolet light source emits a 254nm wavelength beam that penetrates the water sample, and the beam attenuation intensity is detected. According to the Lambert-Beer law, the chemical oxygen demand data in the water body at the fluorine chemical industrial park's discharge outlet are collected. The fluoride ion selective electrode sensor's electrode surface reacts with the fluoride ions in the water sample using a potential response, and the fluoride ion concentration data is obtained through the Nernst equation. For the hourly collected chemical oxygen demand (COD) and fluoride ion concentration (FOC) data, outliers were identified using the 3σ criterion. If a data point exceeded the mean ± 3 times the standard deviation, it was marked as invalid data and removed. For data with no more than 2 consecutive missing hours, linear interpolation of adjacent time points was used to fill the gaps. The collected COD and FOC data were scaled to the [0,1] interval using Min-Max standardization.

4. The method for collaborative verification of PFAS in fluorochemical industrial parks based on dynamic multi-media allocation according to claim 3, characterized in that: In step S1, the process of sampling and detecting the concentration of PFAS in water, sediment, and aquatic organisms includes: Downstream of the fluorine chemical industrial park's discharge outlet, fully automated water quality samplers, gravity column sediment samplers, and trawl-type biological samplers are deployed. Each month, 1 liter of water, 500 grams of sediment, and 50-100 grams of individual aquatic organisms are collected. PFAS are detected using liquid chromatography-mass spectrometry. Solid-phase extraction columns and ion sources are provided. Before detection, water samples are fixed with nitric acid, sediments are extracted ultrasonically with acetonitrile, and biological samples are extracted with accelerated solvents. In mass spectrometry, water samples, sediment samples, and aquatic organism samples are separated by liquid chromatography and then enter an electrospray ionization source to generate PFAS precursor ions in positive ion mode. Using multiple reaction monitoring mode, a specific mass-to-charge ratio is set, and the intensity of PFAS precursor ions is recorded by mass spectrometer. The concentration of PFAS in water, sediment, and aquatic organisms is calibrated based on the internal standard method.

5. The method for collaborative verification of PFAS in fluorochemical industrial parks based on dynamic multi-media allocation according to claim 4, characterized in that: In step S2, the process of extracting fluorochemical characteristics based on preprocessed chemical oxygen demand (COD) and fluoride ion concentration (FOC) data includes: The characteristics of fluorochemicals include the dynamic cumulative index of chemical oxygen demand and the volatility of fluoride ion concentration; Based on the pre-processed hourly chemical oxygen demand (COD) data, with a 24-hour sliding window period, the arithmetic mean and standard deviation of the COD data within the window are calculated. The mean is added to twice the standard deviation to obtain the dynamic cumulative index of COD. The mean and standard deviation of the pretreated fluoride ion concentration data are statistically analyzed monthly, and the fluoride ion concentration fluctuation rate is obtained by using the coefficient of variation formula.

6. The method for collaborative verification of PFAS in fluorochemical industrial parks based on dynamic multi-media allocation according to claim 5, characterized in that: In step S3, the process of constructing a cross-media correlation model to predict the concentration of PFAS in water includes: The extracted fluorochemical features are input into the input layer of the cross-media correlation model, and seasonal coding and real-time flow data are introduced. The hidden layer neurons of the cross-media association model capture the dynamic correlation between fluoride chemical features and PFAS concentration in water bodies, and output the hidden state. Seasonal coding and real-time flow data are fused with the hidden state through a fully connected layer to output the predicted value of PFAS concentration in water bodies. The predicted value of PFAS concentration in water bodies is constrained according to the multi-media distribution law of water-sediment-organism.

7. The method for collaborative verification of PFAS in fluorochemical industrial parks based on dynamic multi-media allocation according to claim 6, characterized in that: S4 specifically includes: Based on the contribution of water bodies, sediments, and aquatic organisms to ecological risks in historical data, we analyze historical data of equal time lengths and combine them with the average PFAS concentrations of various components in water bodies, sediments, and aquatic organisms to determine reference values ​​for PFAS concentrations in water bodies, sediments, and aquatic organisms. The predicted PFAS concentration values ​​for each medium are compared with their corresponding reference values. The deviation rate of the PFAS concentration for each medium is analyzed to determine whether there are any anomalies in the predicted PFAS concentration values ​​for each medium, thereby triggering independent early warnings for each medium.

8. The method for collaborative verification of PFAS in fluorochemical industrial parks based on dynamic multi-media allocation according to claim 7, characterized in that: In step S5, the process of visually displaying the spatial distribution of PFAS concentration and comparing predicted and actual detection values ​​includes: Based on the coordinates of monitoring points in the fluorochemical industrial park and the predicted PFAS concentrations of corresponding water bodies, sediments, and aquatic organisms, the inverse distance weighting method is adopted. A hydrodynamic model is introduced to simulate the pollutant diffusion path. Combining the concentration field output by the hydrodynamic model with the inverse distance weighting interpolation, the discrete point concentration data is interpolated into a continuous spatial distribution. The interpolated concentration is then mapped to color intensity according to the gradient, with red representing high concentration, blue representing medium concentration, and green representing low concentration, thus generating a heat map of the spatial distribution of PFAS concentration in the industrial park. Line charts are used to compare historical forecasts with actual detection data. The comprehensive risk index and actual detection values ​​are aligned by the monthly time axis, as well as the forecast and actual values ​​of water bodies, sediments and aquatic organisms. The monthly forecast values ​​are connected by dashed lines, the horizontal axis is time, the vertical axis is concentration, and the sampling detection values ​​are connected by solid lines. The charts are overlaid with the forecast curves.

Images