Environmental Pollution Monitoring Method and Device Based on Evaluation Model

Abstract

The present invention discloses an environmental pollution monitoring method and device based on an evaluation model, relating to the field of environmental pollution monitoring. The method includes: analyzing the environmental pollution evaluation features of the evaluation model and establishing a multi-dimensional feature list of the evaluation model; analyzing the existence form, diffusion mode, pollution target, and pollution property of the environmental pollution poison monitoring target to obtain the pollution poison monitoring features, and performing a monitoring evaluation target response analysis to obtain the response evaluation demand parameters; using the response evaluation demand parameters to perform a traversal match with the model evaluation features to obtain the missing demand parameters; performing additional training on the evaluation model to obtain an enhanced evaluation model; using the enhanced evaluation model to perform pollutant monitoring and evaluation on environmental monitoring data to obtain the environmental pollution monitoring and evaluation results, and performing pollution monitoring feedback. It solves the technical problem of insufficient accuracy and reliability of the evaluation results in existing environmental pollution monitoring, and achieves the technical effect of improving the accuracy and reliability of the evaluation results.

Description

Technical Field

[0001] This application relates to the field of environmental pollution monitoring, and in particular to an environmental pollution monitoring method and device based on an evaluation model. Background Art

[0002] As one of the major challenges currently faced globally, the accuracy of monitoring and evaluation of environmental pollution is directly related to the effective implementation of environmental protection measures and the maintenance of ecological balance. Currently, the monitoring of environmental pollution mainly relies on some statistical prediction models. These methods evaluate the environmental pollution situation by establishing a prediction model based on historical data. However, although statistical prediction models can handle a large amount of data, they are often limited by the fixed structure of the model and the limitations of training data, and it is difficult to adapt to complex and changeable environmental pollution situations, resulting in insufficient accuracy and reliability of the evaluation results.

[0003] In the current related technologies, there are technical problems of insufficient accuracy and reliability of the evaluation results in environmental pollution monitoring based on an evaluation model. Summary of the Invention

[0004] This application provides an environmental pollution monitoring method and device based on an evaluation model. By using technical means such as analyzing the evaluation model, establishing a multi-dimensional feature list, analyzing the characteristics of pollution poisons, obtaining response evaluation demand parameters, matching the demand parameters with the model characteristics, determining missing parameters, newly training the evaluation model to obtain an enhanced model, and using the enhanced model to evaluate environmental monitoring data, obtaining a monitoring result and giving feedback, etc., the technical effect of improving the accuracy and reliability of the evaluation result is achieved.

[0005] This application provides an environmental pollution monitoring method based on an evaluation model, including: analyzing the environmental pollution evaluation characteristics of the evaluation model and establishing a multi-dimensional feature list of the evaluation model; analyzing the existence form, diffusion mode, pollution target, and pollution property of the environmental pollution poison monitoring target to obtain pollution poison monitoring characteristics, and performing a monitoring evaluation target response analysis according to the pollution poison monitoring characteristics to obtain response evaluation demand parameters; using the response evaluation demand parameters to traverse and match with the model evaluation characteristics in the multi-dimensional feature list of the evaluation model to obtain missing demand parameters; newly training the evaluation model according to the missing demand parameters to obtain an enhanced evaluation model; using the enhanced evaluation model to perform pollutant monitoring and evaluation on environmental monitoring data to obtain an environmental pollution monitoring and evaluation result, and giving pollution monitoring feedback on the environmental pollution monitoring and evaluation result.

[0006] In a possible implementation, for the environmental pollution evaluation features of the parsing evaluation model, a multi-dimensional feature list of the evaluation model is established, and the following processing is performed: The evaluation model is parsed for features in multiple dimensions including training sample features, model architecture features, training methods, output forms, and model performance features to obtain multi-dimensional evaluation features; according to the dimensional feature labels of the multi-dimensional evaluation features, the multi-dimensional evaluation features are integrated to construct the multi-dimensional feature list of the evaluation model.

[0007] In a possible implementation, for the evaluation model, features are parsed in multiple dimensions including training sample features, model architecture features, training methods, output forms, and model performance features to obtain multi-dimensional evaluation features, and the following processing is performed: The training samples are parsed for features such as data source, types of sample pollution poisons, data features, sample quantity distribution, and data labels to obtain polluted training sample features; the pollutant evaluation logic features are analyzed based on the model type and model structure of the model architecture to obtain model architecture features; the training method features are analyzed according to the training type and training parameters to obtain training method features; the output form features are analyzed according to the type, dimension, and range of the model output to obtain output form features; the performance features are analyzed according to the accuracy, real-time performance, and robustness of the model output results to obtain model performance features; the polluted training sample features, model architecture features, training method features, output form features, and model performance features are combined to obtain the multi-dimensional evaluation features.

[0008] In a possible implementation, before performing new training on the evaluation model according to the missing required parameters, the following processing is performed: Determine whether the missing required parameters exist; when they do not exist, use the evaluation model to perform pollutant monitoring and evaluation on environmental monitoring data; when they exist, perform training method analysis based on the missing required parameters and the evaluation model to obtain a new training method; perform new training on the evaluation model based on the missing required parameters and the new training method.

[0009] In a possible implementation, the following processing is performed: The new training method includes transfer learning, incremental learning, or ensemble learning.

[0010] In a possible implementation, the presence form, diffusion mode, pollution target, and pollution analysis of the environmental pollution poison monitoring target are carried out to obtain the pollution poison monitoring characteristics, and the following processing is performed: According to the environmental pollution poison monitoring target, sample data of the pollution target is collected, including historical pollution record data and experimental data, for recording the monitoring concentration, presence form, propagation change characteristics, pollution target impact parts, and pollution impact data of the environmental pollution poison; According to the presence form and the pollution target impact parts, the monitoring characteristics of the collected data are determined; According to the sample data of the pollution target, the spatio-temporal change influence relationship between the propagation change characteristics and the monitoring concentration and pollution impact data is fitted to determine the poison diffusion monitoring characteristics; According to the monitoring characteristics of the collected data and the poison diffusion monitoring characteristics, the pollution poison monitoring characteristics are obtained.

[0011] In a possible implementation, according to the pollution poison monitoring characteristics, a monitoring evaluation target response analysis is carried out to obtain response evaluation requirement parameters, and the following processing is performed: According to the monitoring constraint conditions of the pollution poison, the accuracy relationship fitting of the monitoring characteristics of the collected data and the diffusion path range fitting of the poison diffusion monitoring characteristics are carried out on the pollution poison monitoring characteristics to obtain the monitoring response constraint of the pollution poison monitoring characteristics; Taking the monitoring response constraint of the pollution poison monitoring characteristics as the target, the pollution identification process of the pollution poison monitoring characteristics is analyzed to obtain the response evaluation requirement parameters.

[0012] In a possible implementation, after obtaining the enhanced evaluation model, the following processing is performed: Establish the identification label of the enhanced evaluation model and the input layer screening processing rule; Obtain environmental monitoring data and target identification labels; According to the target identification label and the enhanced evaluation model, matching identification is carried out to obtain a matching model or a matching integrated sub-model; According to the input layer screening processing rule, the environmental monitoring data is screened and processed, and the screened environmental monitoring data is input into the matching model or the matching integrated sub-model.

[0013] In a possible implementation, for obtaining the missing requirement parameters, the following processing is performed: The response evaluation requirement parameters and the model evaluation characteristics in the multi-dimensional feature list are subjected to data standardization and feature dimension alignment processing; Matching rules are configured, including data matching, range matching, and logical matching, and based on the matching rules, the missing requirement parameters and the model evaluation characteristics in the multi-dimensional feature list are matched item by item to obtain the missing parameters and the missing types; According to the missing parameters and the missing types, the missing requirement parameters are obtained.

[0014] The present application also provides an environmental pollution monitoring device based on an evaluation model, including: an environmental pollution evaluation feature analysis module for analyzing the environmental pollution evaluation features of the evaluation model and establishing a multi-dimensional feature list of the evaluation model; a response evaluation requirement parameter acquisition module for analyzing the existence form, diffusion mode, pollution target, and pollution property of the environmental pollution poison monitoring target to obtain pollution poison monitoring features, and performing a monitoring evaluation target response analysis based on the pollution poison monitoring features to obtain response evaluation requirement parameters; a missing requirement parameter acquisition module for traversing and matching the response evaluation requirement parameters with the model evaluation features in the multi-dimensional feature list of the evaluation model to obtain missing requirement parameters; a model new training module for newly training the evaluation model according to the missing requirement parameters to obtain an enhanced evaluation model; and a pollutant monitoring evaluation module for using the enhanced evaluation model to perform pollutant monitoring and evaluation on environmental monitoring data, obtaining an environmental pollution monitoring and evaluation result, and feeding back the environmental pollution monitoring and evaluation result for pollution monitoring.

[0015] It is intended to first analyze the environmental pollution evaluation features of the evaluation model through the environmental pollution monitoring method and device based on the evaluation model proposed in the present application, establish a multi-dimensional feature list of the evaluation model, then analyze the existence form, diffusion mode, pollution target, and pollution property of the environmental pollution poison monitoring target to obtain pollution poison monitoring features, and perform a monitoring evaluation target response analysis based on the pollution poison monitoring features to obtain response evaluation requirement parameters. Then, traverse and match the response evaluation requirement parameters with the model evaluation features in the multi-dimensional feature list of the evaluation model to obtain missing requirement parameters. Next, newly train the evaluation model according to the missing requirement parameters to obtain an enhanced evaluation model. Finally, use the enhanced evaluation model to perform pollutant monitoring and evaluation on environmental monitoring data, obtain an environmental pollution monitoring and evaluation result, and feed back the environmental pollution monitoring and evaluation result for pollution monitoring. The technical effect of improving the accuracy and reliability of the evaluation result is achieved. Brief Description of the Drawings

[0016] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings of the embodiments of the present invention will be briefly introduced below. Flowcharts are used in the present application to illustrate the operations performed by the device according to the embodiments of the present application. It should be understood that the operations in the front or below do not necessarily need to be executed precisely in sequence. On the contrary, according to the need, they can be executed in reverse order or simultaneously. At the same time, other operations can also be added to these processes, or one or several operations can be removed from these processes.

[0017] Figure 1 It is a schematic flowchart of the environmental pollution monitoring method based on the evaluation model provided by the embodiment of the present application.

[0018] Figure 2 The structural schematic diagram of the environmental pollution monitoring device based on the evaluation model provided by the embodiment of the present application.

[0019] Explanation of the reference numerals in the drawings: Environmental pollution evaluation feature analysis module 10, response evaluation requirement parameter acquisition module 20, missing requirement parameter acquisition module 30, model new training module 40, pollutant monitoring evaluation module 50. Detailed implementation manners

[0020] The above description is only an overview of the technical solution of the present application. In order to be able to understand the technical means of the present application more clearly, it can be implemented according to the content of the description. And in order to make the above and other purposes, features and advantages of the present application more obvious and understandable, the following specifically gives the detailed implementation manners of the present application.

[0021] In order to make the purpose, technical solution and advantages of the present application clearer, the present application will be further described in detail below with reference to the drawings. The described embodiments should not be regarded as limitations of the present application. All other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope of protection of the present application.

[0022] In the following description, reference is made to "some embodiments", which describe a subset of all possible embodiments. However, it can be understood that "some embodiments" can be the same subset or different subsets of all possible embodiments, and can be combined with each other without conflict. The terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, device, product or server that includes a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or are inherent to these processes, methods, products or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which the present application belongs. The terms used herein are only for the purpose of describing the embodiments of the present application.

[0023] The embodiment of the present application provides an environmental pollution monitoring method based on an evaluation model, as Figure 1 shown, the method includes:

[0024] Step S100, analyze the environmental pollution evaluation features of the evaluation model, and establish a multi-dimensional feature list of the evaluation model.

[0025] Specifically, obtain an existing environmental pollution assessment model, which is an algorithm model based on machine learning and is used to evaluate the degree or impact of environmental pollution. By reading the model's documentation or source code, or through the feature engineering steps during model training, identify the key features used by the model to evaluate environmental pollution. These features include the types and concentrations of pollutants, the locations of pollution sources, wind direction and speed, etc. Organize the identified evaluation features into a multi-dimensional feature list, with each feature corresponding to a dimension for subsequent matching with the monitoring target features.

[0026] In a possible implementation, for the step S100 of parsing the environmental pollution assessment features of the evaluation model and establishing a multi-dimensional feature list of the evaluation model, it further includes step S110 of performing feature parsing on the evaluation model in multiple dimensions according to training sample features, model architecture features, training methods, output forms, and model performance features to obtain multi-dimensional evaluation features. Specifically, obtain the original data set used to train the evaluation model from the model developer or data provider. Preprocess the original data, including data cleaning (removing duplicates, missing values, or outliers), data transformation (normalization, standardization), etc. Extract the key features from the preprocessed data, which are part of the model input, such as pollutant concentration, timestamp, geographical location, etc. Use statistical methods (such as mean, variance, correlation analysis) or visualization tools (such as scatter plots, histograms) to analyze the distribution and correlation of these features in the training samples.

[0027] Consult the model's documentation or source code to understand the model's architecture, including the structures of the input layer, hidden layer, and output layer, as well as the connection methods between the layers. Use tools (such as TensorBoard, Netron) to visualize the model architecture and analyze the number, type, and initialization methods of the parameters (such as weights, bias terms) in the model, and how they affect the model's performance.

[0028] Identify the algorithms used to train the model from the model documentation or source code, such as gradient descent, Adam optimizer, etc. Analyze the key steps in the training process, such as the adjustment of the learning rate, the selection of the batch size, the application of regularization methods, etc. Review the logs during the training process to understand the performance and performance changes of the model at different training stages.

[0029] Determine the data type of the model output, such as continuous values (pollutant concentration prediction), classification labels (pollution level classification), etc. Analyze the output format of the model, such as JSON, CSV, API response, etc.

[0030] Select appropriate evaluation metrics according to the application scenarios and objectives of the model, such as accuracy, recall, F1-score, mean squared error, etc. Evaluate the performance of the model on the validation set or test set, and use the selected evaluation metrics to measure the accuracy and generalization ability of the model. Compare the performance of the model with the baseline model or other relevant models to understand the advantages and disadvantages of the model and the room for improvement.

[0031] In step S120, integrate the multi-dimensional evaluation features according to the dimension feature labels of the multi-dimensional evaluation features to construct a multi-dimensional feature list of the evaluation model. Specifically, assign a unique label to each multi-dimensional evaluation feature. These labels can be descriptive names or codes for reference and identification. Through manual sorting or using automated tools (such as Excel, Python scripts), classify features with the same or similar attributes together to form a structured multi-dimensional feature list. This list contains detailed descriptions, data types, value ranges, sources (such as training data, model architecture, training methods, etc.) of each feature. This implementation method provides an accurate and comprehensive information basis for subsequent steps by comprehensively parsing and integrating the features of the evaluation model, which helps to quickly identify and utilize these features in subsequent steps, thereby improving the accuracy and efficiency of the environmental pollution monitoring method.

[0032] In a possible implementation, perform feature parsing on the evaluation model in terms of multi-dimensions of training sample features, model architecture features, training methods, output forms, and model performance features to obtain multi-dimensional evaluation features. Step S110 further includes step S111, perform feature parsing on the training samples in terms of data source, sample pollution poison type, data features, sample quantity distribution, and data labels to obtain contaminated training sample features. Specifically, review the metadata or documents of the training samples to confirm the data source channels (such as monitoring stations, laboratories, public databases, etc.). Traverse the training samples, and identify different pollution poison types according to the chemical substance identifiers or classification labels in the samples. Use feature engineering methods (such as statistic calculation, feature selection, feature transformation, etc.) to extract key features from the original data. Count the number of samples of various pollution poisons, and draw a sample quantity distribution map. Use a text editor or data visualization tool (such as Excel, the Pandas library of Python, etc.) to open the file containing the sample data and its labels, view the content of the label column, and determine what the label corresponding to each sample is. Or write a script (such as a Python script) to read the data file and extract the content of the label column, and use the script to perform statistical analysis on the labels, such as calculating the quantity and frequency of each label, and output the types and distributions of the labels.

[0033] Step S112: Analyze the pollutant assessment logic features according to the model type and model structure of the model architecture to obtain the model architecture features. Specifically, consult the model documentation or source code to confirm which type of model it belongs to (such as neural network, decision tree, support vector machine, etc.). Use the model visualization tools or APIs provided by machine learning frameworks (such as TensorFlow, PyTorch, etc.) to generate the structure diagram of the model, showing the input layer, hidden layer, output layer and their connection methods. According to the model structure and training sample features, deduce how the model evaluates pollutants based on the input data, that is, determine the rules and processes for the model to evaluate pollutants according to the input data.

[0034] Step S113: Analyze the training method features according to the training type and training parameters to obtain the training method features. Specifically, consult the model training documentation or code to confirm the training type (such as supervised learning, unsupervised learning, reinforcement learning, etc.). List and review the key parameters used in the training process (such as learning rate, batch size, number of iterations, etc.).

[0035] Step S114: Analyze the output form features according to the type, dimension, and range of the model output to obtain the output form features. Specifically, consult the model documentation or code to confirm the data type of the model output (such as numerical value, classification label, probability distribution, etc.), count the dimension of the model output (such as a single numerical value, vector, matrix, etc.), and determine the data range or boundary of the model output (such as a pollution index from 0 to 100).

[0036] Step S115: Analyze the performance features according to the accuracy, real-time performance, and robustness of the model output results to obtain the model performance features. Specifically, use the test set or validation set to evaluate the accuracy of the model (such as classification accuracy, regression error, etc.). Measure the response time of the model from input to output to evaluate its real-time performance. Analyze the stability and robustness of the model by introducing noise, outliers, or changing the distribution of the input data, etc.

[0037] Step S116: Combine the pollution training sample features, model architecture features, training method features, output form features, and model performance features to obtain the multi-dimensional evaluation features. Specifically, organize, summarize, and combine all the features obtained in Steps S111 to S115 to form a comprehensive multi-dimensional evaluation feature set. This implementation method helps to quickly identify and utilize these features in subsequent steps by combining these features into a comprehensive multi-dimensional evaluation feature set. This structured feature representation method helps with integration and interoperability with other systems or methods.

[0038] Step S200: Analyze the existing form, diffusion mode, pollution target, and pollution characteristics of the environmental pollution toxicant monitoring target to obtain the pollution toxicant monitoring characteristics, and perform monitoring evaluation target response analysis based on the pollution toxicant monitoring characteristics to obtain response evaluation requirement parameters.

[0039] Specifically, through laboratory analysis, such as using techniques like microscope observation, X-ray diffraction (XRD), or scanning electron microscopy (SEM), determine the physical form of the pollution toxicant (such as gas, liquid, solid). For gases, gas chromatography (GC) or mass spectrometry (MS) techniques can be used for analysis; for liquids and solids, techniques such as infrared spectroscopy (IR), nuclear magnetic resonance (NMR), or Raman spectroscopy can be adopted. Use chemical analysis techniques, such as titration, spectrophotometry, or electrochemical methods, to determine the chemical properties of the pollution toxicant, such as volatility and solubility. In addition, by measuring its physical properties such as melting point, boiling point, and density, its chemical stability can also be indirectly reflected.

[0040] Study the diffusion mode of the pollution toxicant in the environment, such as air transmission, water flow transmission, soil penetration, etc. For air transmission, use gas chromatography-mass spectrometry (GC-MS) technology to monitor the concentration of pollutants in the atmosphere, and combine meteorological data (such as wind speed and wind direction) and geographic information system (GIS) to simulate the diffusion path of pollutants. For water flow transmission, collect water samples through water quality monitoring stations, and use techniques such as liquid chromatography-mass spectrometry (LC-MS) or inductively coupled plasma mass spectrometry (ICP-MS) to analyze the types and concentrations of pollutants in water. Combine hydrological models (such as SWAT, MIKE, etc.) to simulate the water flow transmission process of pollutants. For soil penetration, adopt soil sampling and laboratory analysis (such as soil gas extraction-gas chromatography) to evaluate the penetration of pollutants in the soil. At the same time, use soil erosion and sedimentation models (such as RUSLE, WEPP, etc.) to predict the migration and accumulation of pollutants in the soil.

[0041] Identify the objects that the pollution toxicant may affect, such as human health, ecological environment, water resources, etc. For human health, evaluate the impact of pollutants on human health through epidemiological investigations, biological monitoring (such as the content of pollutants in blood and urine), and risk assessment models (such as the IRIS database of USEPA). For the ecological environment, use ecotoxicology experiments (such as acute toxicity tests, chronic toxicity tests) and ecological risk assessment methods (such as the ecological risk assessment framework of OECD) to evaluate the potential impact of pollutants on the ecosystem (such as plants, animals, microorganisms). For water resources, identify the impact of pollutants on water resources through water quality monitoring and ecological health assessment (such as biodiversity index, water quality comprehensive index).

[0042] Evaluate the toxicity, persistence, bioaccumulation, etc. of polluting poisons to determine their potential hazards to the environment and human body. Use toxicity tests (such as acute toxicity tests, genotoxicity tests, carcinogenicity tests) and bioaccumulation experiments (such as the determination of bioconcentration factor BCF and biomagnification factor BMF) to evaluate the toxicity, persistence and bioaccumulation of pollutants. Combine environmental persistence assessment models (such as PBTK model) and biogeochemical cycle models to predict the long-term behavior and impact of pollutants in the environment.

[0043] Based on the above analysis, determine the objectives of monitoring and assessment, such as monitoring the concentration changes of pollutants, evaluating the diffusion range of pollutants, predicting the long-term impact of pollutants, etc. According to the monitoring and assessment objectives, determine the required monitoring parameters, such as monitoring frequency (regular monitoring, real-time monitoring), monitoring location (near pollution sources, sensitive areas), monitoring methods (on-site monitoring, laboratory analysis), etc.

[0044] In a possible implementation manner, the analysis of the existence form, diffusion mode, pollution target, and pollution property of the monitoring target of environmental pollution poisons to obtain the monitoring characteristics of pollution poisons. Step S200 further includes step S210. According to the monitoring target of environmental pollution poisons, collect sample data of pollution targets, including historical pollution record data and experimental data, for recording the monitoring concentration, existence form, propagation change characteristics, pollution target impact parts, and pollution impact data of environmental pollution poisons. Specifically, collect data of past similar environmental pollution events by referring to the historical records of environmental protection departments, the databases of scientific research institutions, or public environmental pollution reports. These data contain information such as the types, concentrations, occurrence times, locations, and influence ranges of pollution poisons. Simulate the environmental pollution scenario under laboratory conditions and record the behavioral characteristics of pollution poisons under different conditions, such as existence form (gaseous, liquid, solid), propagation speed, etc. The experimental data are obtained through chemical analysis, biological testing and other means. The historical pollution record data provide real-world pollution cases, and the experimental data provide the behavioral characteristics of pollutants under controlled conditions. The two are combined to reflect the actual situation of pollutants.

[0045] Step S220, determine the monitoring characteristics of the collected data according to the existence form and the pollution target impact parts. Specifically, analyze the existence form of pollution poisons, such as gaseous, liquid or solid, by chemical analysis means based on the collected sample data. Identify the main parts affected by pollution, such as soil, water source, air or organisms, by comparing the environmental changes before and after the pollution occurs. Accordingly, determine the monitoring characteristics of the collected data, that is, determine the focus and direction of monitoring. Different existence forms and impact parts require different monitoring methods and means.

[0046] Step S230: According to the sample data of the pollution target, fit the spatio-temporal variation influence relationship between the propagation change characteristics, the monitoring concentration, and the pollution impact data to determine the monitoring characteristics of toxicant diffusion. Specifically, using a mathematical model or simulation software, based on information such as the propagation speed and influence range in the sample data, fit the propagation change characteristics of the pollutant (the characteristics of the propagation speed and influence range of the pollutant changing with time in the environment), which can be a dynamic model in time-space to describe the concentration distribution of the pollutant at different times and locations. By analyzing the toxic effects, bioaccumulation conditions, etc. of the pollutant at different concentrations, establish a quantitative relationship between the monitoring concentration (the actual concentration of the pollutant in the environment) and the pollution impact. According to the diffusion law and influence range of the pollutant, determine the monitoring characteristics of toxicant diffusion for formulating effective monitoring strategies.

[0047] Step S240: According to the monitoring characteristics of the collected data and the monitoring characteristics of toxicant diffusion, obtain the monitoring characteristics of the pollution toxicant. Specifically, comprehensively analyze and integrate the monitoring characteristics of the collected data and the monitoring characteristics of toxicant diffusion determined in steps S220 and S230 to form a complete set of monitoring characteristics of the pollution toxicant. This set of characteristics should include information such as the existence form of the pollutant, the affected parts, the propagation change characteristics, and the relationship with the monitoring concentration and pollution impact. This implementation method comprehensively understands the behavior characteristics and potential risks of the pollutant by integrating monitoring characteristics from different aspects, providing a basis for formulating targeted monitoring and prevention and control measures.

[0048] In a possible implementation manner, perform a monitoring evaluation target response analysis according to the monitoring characteristics of the pollution toxicant to obtain response evaluation requirement parameters. Step S200 further includes step S250: According to the monitoring constraint conditions of the pollution toxicant, fit the precision relationship of the monitoring characteristics of the collected data and fit the diffusion path range of the monitoring characteristics of toxicant diffusion for the monitoring characteristics of the pollution toxicant to obtain the monitoring response constraints of the monitoring characteristics of the pollution toxicant. Specifically, analyze the data sources required for the pollution toxicant monitoring target, such as historical pollution records, real-time monitoring data, laboratory analysis results, etc. According to the toxicity level and potential hazards of the pollution toxicant, determine the requirements for monitoring precision. For example, for highly toxic substances, even small concentrations of pollution require high-precision monitoring. Based on historical data and experimental data, use statistical methods or machine learning algorithms to establish a relationship model between the monitoring concentration and data precision. This model is a regression model used to predict the required monitoring precision at different concentrations. According to the requirements of the toxicity level and monitoring precision, adjust and optimize the precision settings of the monitoring system to ensure that the required monitoring precision is achieved within the critical concentration range.

[0049] Using diffusion models in Geographic Information System (GIS) and environmental science, simulate the diffusion paths and ranges of polluting poisons in the environment. These models perform dynamic simulations based on environmental factors such as wind fields and water currents. According to the simulation results, define the possible diffusion areas of polluting poisons, and adjust and optimize the layout of monitoring points according to actual needs to ensure coverage of all key areas.

[0050] In step S260, taking the monitoring response constraints of the polluting poison monitoring characteristics as the goal, analyze the pollution identification process of the polluting poison monitoring characteristics to obtain the response evaluation requirement parameters. Specifically, based on the monitoring data and the results of the diffusion model, identify the main pollution sources and potential pollution paths. Analyze the propagation process, transformation mechanism, and influencing factors of pollutants, including physical, chemical, and biological processes. According to the results of the pollution identification process, determine the key monitoring indicators, such as pollutant concentration, diffusion speed, influence range, etc. Based on the key indicators and monitoring constraint conditions, calculate the technical parameters required, such as monitoring frequency, number of sampling points, monitoring methods, etc. This implementation method ensures that the monitoring system achieves the required monitoring accuracy and coverage range within the key concentration range and diffusion area through precision relationship fitting and diffusion path range fitting. At the same time, through the analysis of the pollution identification process and the extraction of response evaluation requirement parameters, it provides a scientific basis for the design and optimization of the monitoring system, ensuring that the monitoring system can accurately identify pollution sources, evaluate the degree of pollution, and formulate effective countermeasures. This not only improves the pertinence and effectiveness of the monitoring system but also enhances the accuracy and reliability of environmental pollution monitoring.

[0051] In step S300, use the response evaluation requirement parameters to perform a traversal match with the model evaluation characteristics in the multi-dimensional feature list of the evaluation model to obtain the missing requirement parameters.

[0052] Specifically, compare the response evaluation requirement parameters with each feature in the multi-dimensional feature list of the evaluation model one by one to determine which requirement parameters match the model features and which requirement parameters are missing from the model. Through the traversal match, identify the parameters that are not in the model feature list but are required to achieve the monitoring and evaluation goals, that is, the missing requirement parameters.

[0053] In a possible implementation, the step of obtaining the missing requirement parameters, step S300, further includes step S310 of performing data standardization and feature dimension alignment processing on the response evaluation requirement parameters and the model evaluation features in the multi-dimensional feature list. Specifically, data standardization refers to converting data from different sources and in different formats into a unified format and unit for easy comparison and analysis. Feature dimension alignment means ensuring that the feature dimensions in the multi-dimensional feature list and the response evaluation requirement parameters have the same format, unit, and quantity for easy matching and comparison. For numerical parameters such as detection limit and response time, the units are first unified. For example, the unit of the detection limit is converted from ppm to μg / m³, and the response time is converted from seconds to milliseconds to ensure that all numerical parameters have the same measurement unit. For categorical parameters such as sensor type, enumeration values are used for encoding. For example, the "electrochemical" sensor is encoded as 0, and the "optical" sensor is encoded as 1 for easy computer processing. For composite parameters containing multiple dimensions, such as the input dimension "time series + spatial coordinates", it is split into atomic features. For example, the time series feature is marked as 1, and the spatial coordinate feature is also marked as 1 as two independent feature dimensions.

[0054] If the response evaluation requirement parameters contain dimensions not covered in the multi-dimensional feature list, such as "support for satellite data fusion", corresponding columns need to be added to the multi-dimensional feature list to record this feature. For the feature dimensions that already exist in the multi-dimensional feature list and the response evaluation requirement parameters, ensure that their values have the same format and unit. For example, if the "sensor detection limit" in the multi-dimensional feature list is in the unit of μg / m³, the detection limit in the response evaluation requirement parameters should also be converted to the same unit.

[0055] Step S320, configure matching rules, including data matching, range matching, and logical matching, and perform item-by-item matching on the missing requirement parameters and the model evaluation features in the multi-dimensional feature list based on the matching rules to obtain the missing parameters and the missing types. Specifically, configure the matching rules. The matching rules refer to a set of rules used to determine whether the feature values in the response evaluation requirement parameters match those in the multi-dimensional feature list. Data matching means that when the response evaluation requirement parameter is exactly the same as a certain feature value in the multi-dimensional feature list, it is considered a perfect match; range matching means that when the response evaluation requirement parameter is a range (such as response time < 10 seconds) and the feature value in the multi-dimensional feature list is a specific value, it is determined whether the value falls within the range specified by the requirement parameter; logical matching means that when the response evaluation requirement parameter is a functional requirement (such as "support for dynamic warning threshold"), it is checked whether there is a corresponding functional marker (such as "yes / no") in the multi-dimensional feature list to indicate whether the model has this function.

[0056] Traverse each item in the response evaluation requirement parameter list, and according to the configured matching rules, find the corresponding feature dimension in the multi-dimensional feature list and perform the matching. Record the matching results, including whether it matches, the matching type (exact match, range match, logical match), and any reasons for non-matching.

[0057] Step S330, obtain the missing requirement parameters according to the missing parameters and the missing types. Specifically, according to the matching results of step 320, identify all unmatched response evaluation requirement parameters, and these parameters are the missing parameters. Classify each missing parameter to determine its missing type, including data missing (such as lack of training data for specific scenarios), feature missing (such as the model input not including key features), and performance missing (such as the model or hardware performance not meeting the standard). Record each missing parameter and its missing type in the missing requirement parameter record table. This implementation method ensures the accurate matching between the response evaluation requirement parameters and the multi-dimensional feature list of the evaluation model, thereby identifying the requirements not covered by the model currently (i.e., the missing requirement parameters), providing a clear guiding direction for subsequent model optimization, and ensuring that the model can be continuously iterated and improved to meet more specific requirements.

[0058] Step S400, perform additional training on the evaluation model according to the missing requirement parameters to obtain an enhanced evaluation model.

[0059] Specifically, for the missing requirement parameters, collect relevant environmental monitoring data, such as new pollutant concentration data, new pollution source location data, etc. Perform feature engineering processing on the collected new data to extract features related to the missing requirement parameters. Add the processed new data and its features to the training set of the evaluation model and retrain the model to obtain an enhanced evaluation model that includes the missing requirement parameters.

[0060] In a possible implementation, before performing additional training on the evaluation model according to the missing requirement parameters, step S400 further includes step S410, which determines whether the missing requirement parameters exist. Specifically, after obtaining the missing requirement parameters, the system first checks whether these parameters are empty or meet the preset "existence" conditions (such as non-zero value, non-empty string, etc.). This is achieved through conditional judgment statements (such as if statements) in the programming language, and each missing requirement parameter is checked one by one.

[0061] Step S420, when there is none, use the evaluation model to conduct pollutant monitoring and evaluation on environmental monitoring data. Specifically, if the judgment result shows that there are no missing required parameters, that is, all necessary evaluation features are satisfied, the system directly calls the existing evaluation model. Using environmental monitoring data (such as air quality data, water quality data, etc.) as input, through the processing flow of the model, the environmental pollution monitoring and evaluation results are output, that is, step S500 is no longer executed.

[0062] Step S430, when there is one, analyze the training method according to the missing required parameters and the evaluation model to obtain a new training method. Specifically, when there are missing required parameters, analyze the nature of these parameters (such as data type, value range, etc.) and their relationship with the existing features of the evaluation model. According to the analysis results, select a suitable training method to supplement these missing parameters.

[0063] Step S440, conduct new training on the evaluation model based on the missing required parameters and the new training method. Specifically, according to the new training method determined in step 430, collect or generate training data containing the missing required parameters. Use these data to train or fine-tune the evaluation model to supplement the missing feature parameters. During the training process, it may be necessary to adjust the hyperparameters of the model (such as learning rate, batch size, etc.) to optimize the training effect. After training is completed, use the validation dataset to evaluate the performance of the model to ensure that the newly added parameters do not damage the overall performance of the model. Environmental pollution monitoring is a complex and variable task, and the evaluation model needs to continuously adapt to new environmental conditions and pollutant types. This implementation method can ensure that the evaluation model has sufficient comprehensiveness and accuracy to handle various environmental pollution monitoring scenarios by identifying and supplementing missing required parameters. At the same time, adopting a flexible training method can efficiently update the model, reduce the cost and time of retraining, and thus quickly respond to the challenges of environmental changes.

[0064] In a possible implementation, step S430 further includes step S431, and the new training method includes: transfer learning, incremental learning, or ensemble learning.

[0065] Specifically, analyze the nature of the missing required parameters. First, determine the data type (such as numerical type, categorical type, etc.) and data structure (such as vector, matrix, etc.) of the missing required parameters for selecting a suitable training algorithm. Different algorithms have different adaptabilities to data types. Evaluate the importance of the missing required parameters for improving the model performance and whether it is necessary to supplement these parameters urgently. If a certain parameter is crucial and urgent for the model performance, then a training method that can quickly adapt to new data needs to be adopted.

[0066] Analyze and evaluate the current state of the model. The architecture of the model (such as the number of neural network layers, the number of nodes, etc.) will affect the choice of training method. For example, for deep neural networks, incremental learning is more suitable, which can gradually update the model without changing the network structure. The performance of the current model (such as accuracy, recall, etc.) is also an important factor in choosing the training method. If the model performance is already very high, then a more refined training method (such as fine-tuning in transfer learning) needs to be adopted to avoid performance degradation.

[0067] Evaluate the available training resources. Check whether there is enough new data to train the model. If the data is scarce, transfer learning is a better choice, which can utilize the knowledge in related fields to accelerate the training process. Analyze the available computing resources (such as CPU, GPU, memory, etc.). Some training methods (such as incremental learning in deep learning) require more computing resources to gradually update the model.

[0068] Based on the above analysis, select a suitable new training method. If there is a certain similarity between the missing demand parameters and the existing data or tasks, transfer learning is an effective choice, which allows the model to utilize the knowledge learned in related tasks to accelerate the learning process of new tasks. When new data arrives continuously and the model needs to be updated in real time, incremental learning is a suitable choice, which allows the model to gradually learn new data without retraining the entire model. If the missing demand parameters involve multiple different features or models, ensemble learning methods can be used to combine the prediction results of multiple models to improve the overall performance. This implementation method provides strong support and guidance for the new training of the model by comprehensively considering the nature of the missing demand parameters, evaluating the current state of the model, and the available training resources.

[0069] In a possible implementation, after obtaining the enhanced evaluation model, step S400 further includes step S400a of establishing the recognition labels and input layer screening and processing rules of the enhanced evaluation model. Specifically, according to the application scenario and monitoring target of the enhanced evaluation model, specific recognition labels are established for it. These labels are used to identify key information such as the types of pollutants that the model can handle and the monitoring environment. For example, labels such as "PM2.5 monitoring" and "water quality heavy metal monitoring" can be set for the model. According to the input requirements of the model, screening and processing rules for input data are formulated. These rules include requirements in aspects such as data type, data format, and data range. For example, for a water quality monitoring model, key data such as dissolved oxygen, pH value, and heavy metal content in water need to be screened out and converted into a format that the model can recognize.

[0070] Step S400b: Obtain environmental monitoring data and target recognition tags. Specifically, historical monitoring data is collected or obtained in real time through environmental monitoring devices (such as sensors, drones, etc.). These data cover all aspects of the monitoring target, such as the concentration, existing form, and propagation change characteristics of pollutants. According to the requirements of the monitoring task, corresponding recognition tags are set for the obtained environmental monitoring data. These tags match the model recognition tags established in step S400a to correctly select and use the model.

[0071] Step S400c: Perform matching recognition based on the target recognition tags and the enhanced evaluation model to obtain a matching model or a matching integrated sub-model. Specifically, according to the target recognition tags of the environmental monitoring data, a model or an integrated sub-model that matches it is selected from the established enhanced evaluation model library. The matching process can be achieved by comparing the similarity of tags or keyword matching. In some cases, multiple models need to be integrated to improve the accuracy and robustness of the prediction. These integrated sub-models can be models based on different training data, training methods, or model architectures.

[0072] Step S400d: Screen and process the environmental monitoring data according to the input layer screening and processing rules, and input the screened and processed environmental monitoring data into the matching model or the matching integrated sub-model. Specifically, according to the input layer screening and processing rules formulated in step S400a, the environmental monitoring data is screened and processed, including operations such as data cleaning, format conversion, and data range adjustment. The screened and processed environmental monitoring data is input into the model or integrated sub-model obtained by matching recognition for pollutant monitoring and evaluation. This implementation method ensures the quality and accuracy of the input data by establishing recognition tags and input layer screening and processing rules; by matching recognition and the selection of integrated sub-models, it ensures that the selected model matches the requirements of the monitoring task, thereby improving the accuracy and reliability of the prediction.

[0073] Step S500: Use the enhanced evaluation model to conduct pollutant monitoring and evaluation on the environmental monitoring data, obtain the environmental pollution monitoring and evaluation results, and provide pollution monitoring feedback on the environmental pollution monitoring and evaluation results.

[0074] Specifically, actual environmental monitoring data, such as pollutant concentration, wind direction and wind speed, etc., are input into the enhanced evaluation model (the evaluation model containing missing demand parameters), and the enhanced evaluation model is started to process and analyze the input environmental monitoring data. After the model processes the data, it outputs the environmental pollution monitoring evaluation results, including information such as the concentration of pollutants, the diffusion range, and the potential hazards. The evaluation results are fed back to relevant departments or personnel for taking corresponding environmental protection measures. The embodiment of the present application adopts an analytical evaluation model, establishes a multi-dimensional feature list, analyzes the characteristics of pollution poisons, obtains response evaluation demand parameters, matches the demand parameters with the model features, determines the missing parameters, performs additional training on the evaluation model to obtain an enhanced model, uses the enhanced model to evaluate environmental monitoring data, obtains the monitoring results and feeds them back, etc., achieving the technical effect of improving the accuracy and reliability of the evaluation results.

[0075] The following is an example of obtaining an enhanced evaluation model for monitoring sudden water body heavy metal pollution. The existing evaluation model is a time series prediction model based on LSTM (Long Short-Term Memory Network), which is used to predict the concentration change of mercury ions (Hg²⁺) in the river.

[0076] The characteristics of the training samples are as follows: Data source: Water quality monitoring data of a certain river basin in the past 5 years (sampling frequency: once per hour). Features include: mercury ion concentration (μg / L), dissolved oxygen (DO, mg / L), water temperature (°C), pH value, flow velocity (m / s), upstream industrial wastewater discharge (tons per hour).

[0077] The characteristics of the model architecture are as follows: Input layer: Contains 7 time series features (mercury concentration, DO, water temperature, etc.). Hidden layer: 2 layers of LSTM units, with 128 neurons in each layer. Output layer: Predicts the mercury concentration in the next 2 hours.

[0078] The characteristics of the training method are as follows: Loss function: Mean Squared Error (MSE). Optimizer: Adam (learning rate = 0.001). Training parameters: Batch size = 32, number of iterations = 200, Early Stopping to prevent overfitting.

[0079] The characteristics of pollution poison monitoring are as follows: Existing form: Mercury ions mainly exist in dissolved state (Hg²⁺), and part of them are adsorbed on suspended particles. Diffusion method: Diffuses through river flow, and is significantly affected by water flow velocity and river channel topography. Pollution targets: 5 key monitoring points (A1 - A5) along the river, covering drinking water sources and ecologically sensitive areas. Toxicity analysis: Mercury ions have strong neurotoxicity, and low-concentration changes need to be monitored in real time (threshold: 0.01 μg / L).

[0080] The response evaluation requirement parameters are as follows: Monitoring frequency: Real-time (sampling once every minute). Precision requirement: ±5% (±10% error is allowed when the concentration is lower than 0.01 μg / L). Diffusion path prediction: Combine the output of the hydrological model (SWAT) to predict the pollutant migration trajectory.

[0081] The characteristics of the existing evaluation models are shown in Table 1.

[0082] Table 1: Characteristics of Existing Models

[0083]

[0084] The missing requirement parameters are as follows: Lack of real-time performance: The model has a 30-second delay, which cannot meet the real-time early warning requirements. Insufficient low-concentration detection ability: The detection limit is 0.1 μg / L, which is higher than the safety threshold of 0.01 μg / L. Lack of diffusion path prediction function: The hydrological model is not integrated.

[0085] The newly added training method selects transfer learning to fine-tune the existing LSTM model and adds a diffusion path prediction branch. Specifically: Force the layer parameters to be read-only through trainable=False to protect the weight parameters of the first two layers of the LSTM model from being modified; Add a diffusion prediction head, use a fully connected layer as a transition layer, and use the ReLU activation function to enhance the non-linear expression ability. The output layer corresponds to the pollution probability distribution of 5 monitoring points, which corresponds one by one to the 5 key monitoring points in the river basin. Convert the output into a probability distribution. The optimizer selects the RMSprop adaptive learning rate algorithm (more suitable for small-sample fine-tuning than Adam), and the learning rate is 0.0005, which is used to finely adjust the parameter update step size to balance the convergence speed and accuracy. The loss function selects categorical_crossentropy, which is used to accurately match the probability distribution of the monitoring points.

[0086] In the above text, reference is made to Figure 1 The method for environmental pollution monitoring based on an evaluation model according to an embodiment of the present invention is described in detail. Next, reference will be made to Figure 2 Describe an environmental pollution monitoring device based on an evaluation model according to an embodiment of the present invention.

[0087] The environmental pollution monitoring device based on an evaluation model according to an embodiment of the present invention is used to solve the technical problem of insufficient accuracy and reliability of the evaluation results in the existing technology, and achieves the technical effect of improving the accuracy and reliability of the evaluation results. The environmental pollution monitoring device based on an evaluation model includes: an environmental pollution evaluation feature analysis module 10, a response evaluation requirement parameter acquisition module 20, a missing requirement parameter acquisition module 30, a model new training module 40, and a pollutant monitoring and evaluation module 50.

[0088] The environmental pollution assessment feature analysis module 10 is used to analyze the environmental pollution assessment features of the assessment model and establish a multi-dimensional feature list of the assessment model; the response assessment requirement parameter acquisition module 20 is used to analyze the existence form, diffusion mode, pollution target, and pollution property of the environmental pollution poison monitoring target to obtain the pollution poison monitoring features, and perform a response analysis of the monitoring assessment target according to the pollution poison monitoring features to obtain the response assessment requirement parameters; the missing requirement parameter acquisition module 30 is used to traverse and match the response assessment requirement parameters with the model assessment features in the multi-dimensional feature list of the assessment model to obtain the missing requirement parameters; the model new training module 40 is used to newly train the assessment model according to the missing requirement parameters to obtain an enhanced assessment model; the pollutant monitoring assessment module 50 is used to use the enhanced assessment model to perform pollutant monitoring assessment on the environmental monitoring data, obtain the environmental pollution monitoring assessment result, and feedback the pollution monitoring of the environmental pollution monitoring assessment result.

[0089] Next, the specific configuration of the environmental pollution assessment feature analysis module 10 will be described in detail. As described above, to analyze the environmental pollution assessment features of the assessment model and establish a multi-dimensional feature list of the assessment model, the environmental pollution assessment feature analysis module 10 may further include: a model feature analysis unit for performing feature analysis on the assessment model in multiple dimensions of training sample features, model architecture features, training methods, output forms, and model performance features to obtain multi-dimensional assessment features; a multi-dimensional feature list construction unit for integrating the multi-dimensional assessment features according to the dimension feature labels of the multi-dimensional assessment features to construct the multi-dimensional feature list of the assessment model.

[0090] Among them, the evaluation model is subjected to feature analysis in multiple dimensions including training sample features, model architecture features, training methods, output forms, and model performance features to obtain multi-dimensional evaluation features. The model feature analysis unit may further include: a contaminated training sample feature acquisition subunit for performing feature analysis on the training samples in terms of data source, sample contamination poison type, data features, sample quantity distribution, and data labels to obtain contaminated training sample features; a model architecture feature acquisition subunit for analyzing the pollutant evaluation logic features according to the model type and model structure of the model architecture to obtain model architecture features; a training method feature acquisition subunit for analyzing the training method features according to the training type and training parameters to obtain training method features; an output form feature acquisition subunit for analyzing the output form features according to the type, dimension, and range of the model output to obtain output form features; a model performance feature acquisition subunit for analyzing the performance features according to the accuracy, real-time performance, and robustness of the model output results to obtain model performance features; and a feature combination subunit for combining the contaminated training sample features, model architecture features, training method features, output form features, and model performance features to obtain the multi-dimensional evaluation features.

[0091] Next, the specific configuration of the newly added training module 40 of the model will be described in detail. As described above, before performing new training on the evaluation model according to the missing required parameters, the newly added training module 40 of the model may further include: a judgment unit for judging whether the missing required parameters exist; a pollutant monitoring and evaluation unit for, when they do not exist, using the evaluation model to perform pollutant monitoring and evaluation on environmental monitoring data; a training method analysis unit for, when they exist, analyzing the training method according to the missing required parameters and the evaluation model to obtain a newly added training method; and a newly added training unit for performing new training on the evaluation model based on the missing required parameters and the newly added training method.

[0092] Among them, the training method analysis unit may further include: a newly added training method acquisition subunit for the newly added training method including: transfer learning, incremental learning, or ensemble learning.

[0093] Next, the specific configuration of the response evaluation requirement parameter acquisition module 20 will be described in detail. As described above, through the analysis of the existing form, diffusion mode, pollution target, and pollution characteristics of the environmental pollution poison monitoring target, the pollution poison monitoring characteristics are obtained. The response evaluation requirement parameter acquisition module 20 may further include: a sample data collection unit for collecting sample data of the pollution target according to the environmental pollution poison monitoring target, including historical pollution record data and experimental data, which are used to record the monitoring concentration, existing form, propagation change characteristics, pollution target impact site, and pollution impact data of the environmental pollution poison; a collected data monitoring characteristic determination unit for determining the collected data monitoring characteristics according to the existing form and the pollution target impact site; a poison diffusion monitoring characteristic determination unit for fitting the spatio-temporal change influence relationship between the propagation change characteristics, monitoring concentration, and pollution impact data according to the sample data of the pollution target to determine the poison diffusion monitoring characteristics; and a pollution poison monitoring characteristic acquisition unit for obtaining the pollution poison monitoring characteristics according to the collected data monitoring characteristics and the poison diffusion monitoring characteristics.

[0094] Among them, for the monitoring evaluation target response analysis according to the pollution poison monitoring characteristics to obtain the response evaluation requirement parameters, the response evaluation requirement parameter acquisition module 20 may further include: a monitoring response constraint acquisition unit for fitting the accuracy relationship of the collected data monitoring characteristics and the diffusion path range of the poison diffusion monitoring characteristics to the pollution poison monitoring characteristics according to the monitoring constraint conditions of the pollution poison, so as to obtain the monitoring response constraints of the pollution poison monitoring characteristics; and a pollution identification process analysis unit for analyzing the pollution identification process of the pollution poison monitoring characteristics with the monitoring response constraints of the pollution poison monitoring characteristics as the target to obtain the response evaluation requirement parameters.

[0095] Among them, after obtaining the enhanced evaluation model, the device may further include: an identification label and input layer screening processing rule establishment module for establishing the identification label and input layer screening processing rules of the enhanced evaluation model; an environmental monitoring data and target identification label acquisition module for acquiring environmental monitoring data and target identification labels; a matching identification module for performing matching identification according to the target identification label and the enhanced evaluation model to obtain a matching model or a matching integrated sub-model; and a screening processing module for screening and processing the environmental monitoring data according to the input layer screening processing rules and inputting the screened and processed environmental monitoring data into the matching model or the matching integrated sub-model.

[0096] Next, the specific configuration of the missing requirement parameter acquisition module 30 will be described in detail. As described above, to obtain the missing requirement parameters, the missing requirement parameter acquisition module 30 may further include: a data standardization and feature dimension alignment processing unit for performing data standardization and feature dimension alignment processing on the response evaluation requirement parameters and the model evaluation features in the multi-dimensional feature list; an item-by-item matching unit for configuring matching rules, including data matching, range matching, and logical matching, and performing item-by-item matching on the missing requirement parameters and the model evaluation features in the multi-dimensional feature list based on the matching rules to obtain the missing parameters and the missing types; a missing requirement parameter acquisition unit for obtaining the missing requirement parameters according to the missing parameters and the missing types.

[0097] The environmental pollution monitoring device based on the evaluation model provided by the embodiments of the present invention can execute the environmental pollution monitoring method based on the evaluation model provided by any embodiment of the present invention, and has the corresponding functional modules and beneficial effects for executing the method.

[0098] Although various references are made to certain modules in the devices according to the embodiments of the present application, any number of different modules can be used and run on the user terminal and / or the server. The various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for easy distinction from each other and do not limit the protection scope of the present invention.

[0099] The above specific embodiments do not constitute a limitation on the protection scope of the present application. Those skilled in the art should understand that various modifications, combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principle of the present application should be included within the protection scope of the present application. In some cases, the actions or steps recorded in the present application can be executed in a different order from that in the embodiments and still achieve the desired results. Additionally, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In certain embodiments, multi-tasking and parallel processing are also possible or may be advantageous.

Claims

1. An environmental pollution monitoring method based on an evaluation model, characterized in that, Including: Analyze the environmental pollution assessment features of the evaluation model, and establish a multi-dimensional feature list of the evaluation model, including: Perform feature analysis on the evaluation model in multiple dimensions, including training sample features, model architecture features, training methods, output forms, and model performance features, to obtain multi-dimensional evaluation features, including: Perform feature analysis on the training samples for data source, sample pollution poison type, data features, sample quantity distribution, and data labels to obtain contaminated training sample features; Analyze the pollutant assessment logic features based on the model type and model structure of the model architecture to obtain model architecture features; Analyze the training method features according to the training type and training parameters to obtain training method features; Analyze the output form features based on the type, dimension, and range of the model output to obtain output form features; Analyze the performance features based on the accuracy, real-time performance, and robustness of the model output results to obtain model performance features; Combine the contaminated training sample features, model architecture features, training method features, output form features, and model performance features to obtain the multi-dimensional evaluation features; Integrate the multi-dimensional evaluation features according to the dimension feature labels of the multi-dimensional evaluation features to construct the multi-dimensional feature list of the evaluation model; Analyze the existence form, diffusion method, pollution target, and pollution property of the environmental pollution poison monitoring target to obtain pollution poison monitoring features, and perform monitoring evaluation target response analysis based on the pollution poison monitoring features to obtain response evaluation requirement parameters; Use the response evaluation requirement parameters to traverse and match with the model evaluation features in the multi-dimensional feature list of the evaluation model to obtain missing requirement parameters; Perform additional training on the evaluation model according to the missing requirement parameters to obtain an enhanced evaluation model; Use the enhanced evaluation model to perform pollutant monitoring and evaluation on environmental monitoring data, obtain environmental pollution monitoring and evaluation results, and provide pollution monitoring feedback on the environmental pollution monitoring and evaluation results; After obtaining the enhanced evaluation model, it further includes: Establish the identification label of the enhanced evaluation model and the input layer screening processing rule; Obtain environmental monitoring data and target identification labels; Match and identify according to the target identification label and the enhanced evaluation model to obtain a matching model or a matching integrated sub-model; Screen and process the environmental monitoring data according to the input layer screening processing rule, and input the screened environmental monitoring data into the matching model or the matching integrated sub-model; The obtaining of the missing requirement parameters includes: Perform data standardization and feature dimension alignment processing on the response evaluation requirement parameters and the model evaluation features in the multi-dimensional feature list; Configure matching rules, including data matching, range matching, and logical matching, and perform item-by-item matching on the missing requirement parameters and the model evaluation features in the multi-dimensional feature list based on the matching rules to obtain missing parameters and missing types; Obtain the missing requirement parameters according to the missing parameters and missing types.

2. The environmental pollution monitoring method based on an evaluation model according to claim 1, wherein Before performing additional training on the evaluation model according to the missing requirement parameters, it further includes: Judge whether the missing requirement parameters exist; When it does not exist, use the evaluation model to monitor and evaluate pollutants in environmental monitoring data; When it exists, analyze the training method according to the missing demand parameter and the evaluation model to obtain a new training method; Based on the missing demand parameter and the new training method, perform new training on the evaluation model.

3. The environmental pollution monitoring method based on an evaluation model according to claim 2, characterized in that The new training method includes: transfer learning, incremental learning, or ensemble learning.

4. The environmental pollution monitoring method based on an evaluation model according to claim 1, wherein Analyze the existence form, diffusion method, pollution target, and pollution property of the environmental pollution toxicant monitoring target to obtain pollution toxicant monitoring features, including: According to the environmental pollution toxicant monitoring target, collect sample data of the pollution target, including historical pollution record data and experimental data, for recording the monitoring concentration, existence form, propagation change characteristics, pollution target impact parts, and pollution impact data of the environmental pollution toxicant; According to the existence form and the pollution target impact parts, determine the monitoring features of the collected data; According to the sample data of the pollution target, fit the spatio-temporal variation influence relationship between the propagation change characteristics, monitoring concentration, and pollution impact data to determine the toxicant diffusion monitoring features; According to the monitoring features of the collected data and the toxicant diffusion monitoring features, obtain the pollution toxicant monitoring features.

5. The environmental pollution monitoring method based on an evaluation model according to claim 4, characterized in that Conduct a monitoring evaluation target response analysis according to the pollution toxicant monitoring features to obtain response evaluation demand parameters, including: According to the monitoring constraint conditions of the pollution toxicant, fit the accuracy relationship of the monitoring features of the collected data and the diffusion path range of the toxicant diffusion monitoring features of the pollution toxicant monitoring features to obtain the monitoring response constraint of the pollution toxicant monitoring features; Taking the monitoring response constraint of the pollution toxicant monitoring features as the target, analyze the pollution identification process of the pollution toxicant monitoring features to obtain the response evaluation demand parameters.

6. An environmental pollution monitoring device based on an evaluation model, characterized in that The device is used to implement the environmental pollution monitoring method based on the evaluation model according to any one of claims 1-5. The device includes: An environmental pollution evaluation feature analysis module, used to analyze the environmental pollution evaluation features of the evaluation model and establish a multi-dimensional feature list of the evaluation model; A response evaluation demand parameter acquisition module, used to analyze the existence form, diffusion method, pollution target, and pollution property of the environmental pollution toxicant monitoring target to obtain pollution toxicant monitoring features, and conduct a monitoring evaluation target response analysis according to the pollution toxicant monitoring features to obtain response evaluation demand parameters; A missing demand parameter acquisition module, used to traverse and match the response evaluation demand parameters with the model evaluation features in the multi-dimensional feature list of the evaluation model to obtain missing demand parameters; A model new training module, used to perform new training on the evaluation model according to the missing demand parameters to obtain an enhanced evaluation model; A pollutant monitoring and evaluation module, used to use the enhanced evaluation model to monitor and evaluate pollutants in environmental monitoring data, obtain an environmental pollution monitoring and evaluation result, and feedback the environmental pollution monitoring and evaluation result for pollution monitoring.

Images