Water environment real-time monitoring and early warning method and system based on Internet of Things

Abstract

The invention discloses a water environment real-time monitoring and early warning method and system based on the Internet of Things. The method comprises the following steps: acquiring concentration data and water body attributes; performing feature analysis according to the concentration data to obtain a prediction model, and further performing optimization to obtain optimized parameters; adjusting the prediction model according to the optimization parameters to obtain an optimal prediction model; performing layout analysis according to the concentration data to obtain a monitoring layout scheme; performing threshold analysis according to the water body attributes to obtain a threshold scheme; integrating the optimal prediction model and a threshold scheme to obtain an early warning scheme; adjusting the monitoring layout scheme and the optimization parameters according to the feedback of the early warning scheme in the system to obtain an early warning optimization scheme; and setting the system according to the early warning optimization scheme, so that the system works according to the early warning optimization scheme. According to the invention, the early warning precision and timeliness of the system can be improved by dynamically adjusting the monitoring scheme and the model parameters.

Description

Technical Field

[0001] The present invention relates to the technical field of water environment monitoring and management, and in particular to a real-time water environment monitoring and early warning method and system based on the Internet of Things. Background Art

[0002] Real-time monitoring of dissolved oxygen in aquatic environments presents numerous technical difficulties and contradictions. First, dissolved oxygen concentrations vary significantly across different water layers, with complex and variable vertical and horizontal distributions. A large number of monitoring points and frequent data acquisition are required to fully and accurately capture the spatial and temporal distribution patterns of dissolved oxygen. Second, numerous factors influence dissolved oxygen concentration, such as water velocity, temperature, and the degree of eutrophication. The complex dynamics of dissolved oxygen concentrations, driven by these multiple factors, make it difficult to establish a unified dissolved oxygen distribution model. Furthermore, the significant variation in environmental parameters across different water bodies makes model parameter calibration and applicability verification a significant undertaking. Furthermore, dissolved oxygen concentrations vary over short timescales and large spatial scales, requiring high-speed concentration warning and control, as well as high spatial coverage, placing stringent demands on the deployment and maintenance of monitoring equipment. Finally, determining dissolved oxygen concentration thresholds requires comprehensive consideration of factors such as water body function, environmental capacity, and ecological protection objectives. This involves extensive basic research and coordination between regulatory agencies, making practical implementation challenging. Therefore, a method for determining dissolved oxygen concentration in water is urgently needed.

[0003] In one prior art, a specific implementation includes the following: In prior art, real-time monitoring of dissolved oxygen in aquatic environments typically involves deploying multiple monitoring points within a water body, equipped with dissolved oxygen sensors, and regularly collecting data to understand the spatiotemporal distribution of dissolved oxygen within the water layer. During implementation, these sensors record dissolved oxygen concentrations in real time and upload the data to a central management system for analysis and display, enabling timely detection of abnormalities.

[0004] However, in the existing technology, the layout of sensors and the data acquisition frequency are often insufficient to capture the dynamic changes in dissolved oxygen concentration, resulting in the monitoring results being unable to accurately reflect the actual conditions of the water body. Summary of the Invention

[0005] The present invention provides a real-time water environment monitoring and early warning method and system based on the Internet of Things to solve the problem in the prior art that the layout of sensors and the data acquisition frequency are often insufficient to capture the dynamic changes in dissolved oxygen concentration, resulting in the monitoring results being unable to accurately reflect the actual condition of the water body.

[0006] In the first aspect, in order to solve the above technical problems, the present invention provides a real-time monitoring and early warning method for water environment based on the Internet of Things, comprising:

[0007] Acquiring concentration data and water body attributes; wherein the water body attributes include water body function, environmental capacity, and ecological objectives;

[0008] Performing feature analysis based on the concentration data to obtain a prediction model;

[0009] Optimize according to the prediction model to obtain optimized parameters;

[0010] Adjusting the prediction model according to the optimization parameters to obtain an optimal prediction model;

[0011] Performing layout analysis based on the concentration data to obtain a monitoring layout plan;

[0012] Performing a threshold analysis based on the water body function, the environmental capacity, and the ecological goal to obtain a threshold solution;

[0013] Integrating the optimal prediction model and the threshold scheme to obtain an early warning scheme;

[0014] Adjust the monitoring layout plan and the optimization parameters according to the feedback of the early warning plan in the system to obtain an early warning optimization plan;

[0015] The system is configured according to the early warning optimization plan so that the system works according to the early warning optimization plan.

[0016] In one implementation of the first aspect, performing feature analysis based on the concentration data to obtain a prediction model includes:

[0017] Decomposing the concentration data to obtain trend item data, period item data and random item data;

[0018] Perform long-term fitting based on the trend item data to obtain trend change rules;

[0019] Transform and extract the periodic item data to obtain a periodic variation pattern;

[0020] Perform short-term fitting based on the random item data to obtain random fluctuation patterns;

[0021] A prediction model is constructed based on the trend change law, the period change law and the random fluctuation law.

[0022] In an implementation manner of the first aspect, evaluating according to the prediction model to obtain optimization parameters includes:

[0023] The prediction model is evaluated using a cross-validation method to obtain a cross-validation result;

[0024] Calibration is performed based on the cross-validation results to obtain optimized parameters.

[0025] In an implementation manner of the first aspect, adjusting the prediction model according to the optimization parameters to obtain the optimal prediction model includes:

[0026] Setting the prediction model according to the optimization parameters to obtain an optimized model;

[0027] Perform prediction according to the optimization model to obtain a predicted value;

[0028] Subtracting the predicted value from the pre-stored measured value to obtain a deviation value;

[0029] A judgment is made based on the deviation value and the preset threshold range; if the deviation value is within the preset threshold range, the optimization model is determined to be the optimal prediction model; if the deviation value exceeds the preset threshold range, the optimization model is iteratively optimized using the neural network structure optimization technology to obtain a new model and the above operation is repeated until the optimal prediction model is obtained.

[0030] In one implementation of the first aspect, the iterative optimization of the optimization model using a neural network structure optimization technique to obtain a new model includes:

[0031] Performing structural optimization according to the optimization model to obtain a new model structure;

[0032] Mining is performed based on the relevance and characteristics of the optimization model to obtain influencing factors;

[0033] The optimized model is recalibrated according to the new model structure and the influencing factors to obtain a new model.

[0034] In one implementation of the first aspect, performing layout analysis based on the concentration data to obtain a monitoring layout plan includes:

[0035] Performing position reading on the concentration data to obtain position information;

[0036] Integrate the concentration data and location information to obtain monitoring point data;

[0037] Performing spatial difference analysis based on the monitoring point data to obtain a spatial distribution map;

[0038] Superimposing the spatial distribution map with a pre-stored water body function zoning layer, a pre-stored pollution source distribution layer, and pre-stored hydrological dynamic condition data to obtain a superposition result;

[0039] Identify based on the superposition results to obtain spatial relationships and influencing factors;

[0040] Constructing a model based on the spatial relationship and the influencing factors to obtain a concentration distribution prediction model;

[0041] The temporal and spatial representativeness is optimized based on the concentration distribution prediction model to obtain the monitoring layout plan.

[0042] In an implementation manner of the first aspect, performing a threshold analysis based on the water body function, the environmental capacity, and the ecological target to obtain a threshold solution includes:

[0043] Building a model based on the water body function, the environmental capacity and the ecological goal to obtain a threshold evaluation index;

[0044] Performing index calculation according to the threshold evaluation index to obtain a comprehensive evaluation value;

[0045] An algorithm is used to generate a solution based on the comprehensive evaluation value and pre-stored management requirements to obtain a threshold solution.

[0046] In an implementable manner of the first aspect, the threshold scheme refers to a scheme that can clearly define the warning level, response measures and information disclosure mechanism; for the warning level attribute, the rule engine technology can be used to automatically determine the warning level of the current dissolved oxygen concentration based on the preset IF-THEN rules. If the dissolved oxygen concentration is lower than a certain warning level threshold, the corresponding warning information is triggered; for the response measure attribute, the case reasoning technology is used to retrieve historical cases similar to the current water body situation based on the water body function and the environmental capacity attribute, obtain the response measures taken in the case, and make appropriate adjustments based on the actual situation to form a targeted response measure plan; for the information disclosure attribute, according to the content of the plan, a public template for information such as the dissolved oxygen concentration warning level and response measures is automatically generated for display.

[0047] In a second aspect, the present invention provides a diesel generator set fault prediction model, comprising:

[0048] A data acquisition module, configured to acquire concentration data and water attributes, wherein the water attributes include water function, environmental capacity, and ecological objectives;

[0049] A model building module, used to perform feature analysis based on the concentration data to obtain a prediction model;

[0050] A parameter optimization module, configured to perform optimization based on the prediction model to obtain optimized parameters;

[0051] A model optimization module, configured to adjust the prediction model according to the optimization parameters to obtain an optimal prediction model;

[0052] A layout analysis module, configured to perform layout analysis based on the concentration data to obtain a monitoring layout plan;

[0053] A threshold analysis module, configured to perform a threshold analysis based on the water body function, the environmental capacity, and the ecological target to obtain a threshold solution;

[0054] An integration module, configured to integrate the optimal prediction model and the threshold solution to obtain an early warning solution;

[0055] A feedback adjustment module, configured to adjust the monitoring deployment plan and the optimization parameters according to the feedback of the early warning plan in the system to obtain an early warning optimization plan;

[0056] The system setting module is used to set the system according to the early warning optimization plan so that the system works according to the early warning optimization plan.

[0057] In one implementation of the second aspect, performing feature analysis based on the concentration data to obtain a prediction model includes:

[0058] Decomposing the concentration data to obtain trend item data, period item data and random item data;

[0059] Perform long-term fitting based on the trend item data to obtain trend change rules;

[0060] Transform and extract the periodic item data to obtain a periodic variation pattern;

[0061] Perform short-term fitting based on the random item data to obtain random fluctuation patterns;

[0062] A prediction model is constructed based on the trend change law, the period change law and the random fluctuation law.

[0063] In an implementation manner of the second aspect, evaluating according to the prediction model to obtain optimization parameters includes:

[0064] The prediction model is evaluated using a cross-validation method to obtain a cross-validation result;

[0065] Calibration is performed based on the cross-validation results to obtain optimized parameters.

[0066] In an implementation manner of the second aspect, adjusting the prediction model according to the optimization parameters to obtain the optimal prediction model includes:

[0067] Setting the prediction model according to the optimization parameters to obtain an optimized model;

[0068] Perform prediction according to the optimization model to obtain a predicted value;

[0069] Subtracting the predicted value from the pre-stored measured value to obtain a deviation value;

[0070] A judgment is made based on the deviation value and the preset threshold range; if the deviation value is within the preset threshold range, the optimization model is determined to be the optimal prediction model; if the deviation value exceeds the preset threshold range, the optimization model is iteratively optimized using the neural network structure optimization technology to obtain a new model and the above operation is repeated until the optimal prediction model is obtained.

[0071] In one implementation of the second aspect, the iterative optimization of the optimization model using a neural network structure optimization technique to obtain a new model includes:

[0072] Performing structural optimization according to the optimization model to obtain a new model structure;

[0073] Mining is performed based on the relevance and characteristics of the optimization model to obtain influencing factors;

[0074] The optimized model is recalibrated according to the new model structure and the influencing factors to obtain a new model.

[0075] In an implementation of the second aspect, performing layout analysis based on the concentration data to obtain a monitoring layout plan includes:

[0076] Performing position reading on the concentration data to obtain position information;

[0077] Integrate the concentration data and location information to obtain monitoring point data;

[0078] Performing spatial difference analysis based on the monitoring point data to obtain a spatial distribution map;

[0079] Superimposing the spatial distribution map with a pre-stored water body function zoning layer, a pre-stored pollution source distribution layer, and pre-stored hydrological dynamic condition data to obtain a superposition result;

[0080] Identify based on the superposition results to obtain spatial relationships and influencing factors;

[0081] Constructing a model based on the spatial relationship and the influencing factors to obtain a concentration distribution prediction model;

[0082] The temporal and spatial representativeness is optimized based on the concentration distribution prediction model to obtain the monitoring layout plan.

[0083] In one implementation of the second aspect, performing a threshold analysis based on the water body function, the environmental capacity, and the ecological target to obtain a threshold solution includes:

[0084] Building a model based on the water body function, the environmental capacity and the ecological goal to obtain a threshold evaluation index;

[0085] Performing index calculation according to the threshold evaluation index to obtain a comprehensive evaluation value;

[0086] An algorithm is used to generate a solution based on the comprehensive evaluation value and pre-stored management requirements to obtain a threshold solution.

[0087] In an implementable manner of the second aspect, the threshold scheme refers to a scheme that can clearly define the warning level, response measures and information disclosure mechanism; for the warning level attribute, the rule engine technology can be used to automatically determine the warning level of the current dissolved oxygen concentration based on the preset IF-THEN rules. If the dissolved oxygen concentration is lower than a certain warning level threshold, the corresponding warning information is triggered; for the response measure attribute, the case reasoning technology is used to retrieve historical cases similar to the current water body situation based on the water body function and the environmental capacity attribute, obtain the response measures taken in the case, and make appropriate adjustments based on the actual situation to form a targeted response measure plan; for the information disclosure attribute, according to the content of the plan, a public template for information such as the dissolved oxygen concentration warning level and response measures is automatically generated for display.

[0088] In a third aspect, the present invention also provides an electronic device comprising a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein when the processor executes the computer program, it implements any one of the above-mentioned methods for real-time monitoring and early warning of water environment based on the Internet of Things.

[0089] In a fourth aspect, the present invention also provides a computer-readable storage medium, which includes a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute any one of the above-mentioned real-time water environment monitoring and early warning methods based on the Internet of Things.

[0090] Compared with the prior art, the present invention has the following beneficial effects: the present invention discloses a real-time monitoring and early warning method for water environment based on the Internet of Things, including obtaining concentration data and water body attributes; wherein the water body attributes include water body functions, environmental capacity and ecological targets; performing feature analysis based on the concentration data to obtain a prediction model; performing optimization based on the prediction model to obtain optimization parameters; adjusting the prediction model based on the optimization parameters to obtain an optimal prediction model; performing layout analysis based on the concentration data to obtain a monitoring layout plan; performing threshold analysis based on the water body function, the environmental capacity and the ecological target to obtain a threshold plan; integrating the optimal prediction model and the threshold plan to obtain an early warning plan; adjusting the monitoring layout plan and the optimization parameters based on the feedback of the early warning plan in the system to obtain an early warning optimization plan; setting the system based on the early warning optimization plan so that the system works according to the early warning optimization plan. This method uses a sensor network to collect high-frequency, multi-point dissolved oxygen concentration data. It then integrates factors such as temperature, water velocity, and eutrophication to construct a dissolved oxygen concentration prediction model. Time series analysis and a random forest algorithm are used to process the data and calibrate model parameters. Spatial interpolation is performed using a geographic information system to generate a dissolved oxygen concentration distribution map. Based on water body functions and ecological objectives, dissolved oxygen concentration thresholds are determined to construct a real-time early warning system. By dynamically adjusting monitoring schemes and model parameters, this method continuously optimizes early warning accuracy and timeliness, providing decision-making support for water quality improvement and ecological restoration. This enables precise monitoring, prediction, and early warning of dissolved oxygen concentration in water, enhancing the scientific nature and effectiveness of water environment management. BRIEF DESCRIPTION OF THE DRAWINGS

[0091] Figure 1 This is a flow chart of a method for real-time monitoring and early warning of water environment based on the Internet of Things provided by the first embodiment of the present invention;

[0092] Figure 2 This is a structural diagram of a water environment real-time monitoring and early warning system based on the Internet of Things provided by the second embodiment of the present invention. DETAILED DESCRIPTION

[0093] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0094] Reference Figure 1 The first embodiment of the present invention provides a water environment real-time monitoring and early warning method based on the Internet of Things, comprising the following steps:

[0095] S1, obtaining concentration data and water body attributes; wherein the water body attributes include water body function, environmental capacity and ecological objectives;

[0096] S2, performing feature analysis based on the concentration data to obtain a prediction model;

[0097] S3, optimizing according to the prediction model to obtain optimized parameters;

[0098] S4, adjusting the prediction model according to the optimization parameters to obtain an optimal prediction model;

[0099] S5, performing layout analysis based on the concentration data to obtain a monitoring layout plan;

[0100] S6, performing a threshold analysis based on the water body function, the environmental capacity, and the ecological goal to obtain a threshold solution;

[0101] S7, integrating the optimal prediction model and the threshold scheme to obtain an early warning scheme;

[0102] S8, adjusting the monitoring deployment plan and the optimization parameters according to the feedback of the early warning plan in the system to obtain an early warning optimization plan;

[0103] S9, setting the system according to the early warning optimization plan so that the system works according to the early warning optimization plan.

[0104] In step S1, concentration data and water body attributes are obtained; wherein the water body attributes include water body functions, environmental capacity and ecological goals.

[0105] In a specific embodiment, according to the characteristics of the water environment, appropriate types and quantities of dissolved oxygen sensors are selected, and multiple monitoring points are reasonably arranged to form a sensor network to achieve full coverage monitoring of the target water area. The original data of dissolved oxygen concentration at each monitoring point is obtained through the sensor network, and the data is transmitted to the data processing center. If data is missing or abnormal during the transmission process, the data repair algorithm is started to estimate and correct the missing or abnormal data to ensure the integrity and accuracy of the data. The environmental parameter data such as water temperature and water flow velocity at each monitoring point are obtained, and they are matched with the dissolved oxygen concentration data in time and space to obtain a synchronized multi-parameter coupling data set as concentration data. The water body attributes are information directly retrieved from the website, and the information is imported to form the water body attributes of the system. The water body attributes include water body functions, environmental capacity and ecological goals.

[0106] In step S2, feature analysis is performed based on the concentration data to obtain a prediction model.

[0107] In the above step S2, the feature analysis is performed based on the concentration data to obtain a prediction model, which specifically includes the following steps:

[0108] S21, decomposing the concentration data to obtain trend item data, period item data, and random item data;

[0109] S22, performing long-term fitting based on the trend item data to obtain a trend change rule;

[0110] S23, transforming and extracting the periodic item data to obtain a periodic variation pattern;

[0111] S24, performing short-term fitting based on the random item data to obtain a random fluctuation pattern;

[0112] S25, constructing a prediction model based on the trend change law, the periodic change law and the random fluctuation law.

[0113] It should be noted that the prediction model in the following description refers to a time series prediction model. In the above steps S21 to S25, the specific implementation process includes: using the dissolved oxygen concentration data collected at high frequency, according to the time series characteristics of the data, using the time series decomposition algorithm to decompose the dissolved oxygen concentration data into trend items, periodic items and random items; for the trend item data obtained by decomposition, using the polynomial fitting algorithm to fit the long-term change trend of the dissolved oxygen concentration, and obtain the trend change law of the dissolved oxygen concentration. For the periodic item data obtained by decomposition, using the Fourier transform algorithm to extract the periodic change characteristics of the dissolved oxygen concentration, and obtain the periodic change law of the dissolved oxygen concentration; for the random item data obtained by decomposition, using the autoregressive sliding average model to fit the short-term random fluctuation of the dissolved oxygen concentration, and obtain the random fluctuation law of the dissolved oxygen concentration; according to the trend change law, periodic change law and random fluctuation law of the dissolved oxygen concentration, a time series prediction model for the dissolved oxygen concentration is constructed.

[0114] For example, by decomposing data into trend, cycle, and random terms, we can gain a deeper understanding of the dynamics of dissolved oxygen in water. For example, long-term monitoring data from a particular lake show significant seasonal fluctuations in dissolved oxygen concentration. The trend term reflects the long-term trend in dissolved oxygen concentration, which can be influenced by factors such as climate change and eutrophication. Polynomial fitting can capture this long-term trend. For example, it was found that the dissolved oxygen concentration in this lake exhibits a slow downward trend, with an average annual decrease of 0.1 mg / L. The cycle term reflects the cyclical variation in dissolved oxygen concentration. Using Fourier transforms, we can extract periodic features at different scales. In this lake, distinct annual and daily cycles were observed. The annual cycle is characterized by lower dissolved oxygen concentrations in the summer (approximately 6 mg / L) and higher concentrations in the winter (approximately 10 mg / L), which is closely related to water temperature fluctuations and biological activity. The daily cycle shows an increase in dissolved oxygen concentration during the day and a decrease at night, with a fluctuation of approximately 1-2 mg / L, reflecting the influence of photosynthesis in aquatic plants. The random term represents short-term random fluctuations, which can be influenced by factors such as weather changes and human activities. The autoregressive moving average model can capture these short-term fluctuations, helping to improve forecast accuracy. For example, after a heavy rainfall event, dissolved oxygen concentrations can experience brief fluctuations, with amplitudes as high as 1.5 mg / L. Based on the above analysis, a time series prediction model for dissolved oxygen concentration can be constructed. This model comprehensively considers long-term trends, cyclical changes, and short-term fluctuations, enabling relatively accurate predictions of dissolved oxygen concentrations over the next period of time. For example, the model predicts that the average dissolved oxygen concentration in the lake will be 7.8 mg / L next month, within 0.3 mg / L of the actual observed value. Spatial interpolation is an important tool for understanding the spatial distribution of dissolved oxygen concentrations. The kriging interpolation algorithm accounts for spatial autocorrelation and can provide reasonable estimates between a limited number of sampling points.

[0115] In step S3, optimization is performed according to the prediction model to obtain optimized parameters.

[0116] In the above step S3, the evaluation is performed according to the prediction model to obtain the optimization parameters, which specifically includes the following steps:

[0117] S31, evaluating the prediction model using a cross-validation method to obtain a cross-validation result;

[0118] S32, calibrating according to the cross-validation result to obtain optimized parameters.

[0119] It should be noted that the above steps S31 to S32 specifically include the following implementation process: Based on the prediction model, the model is trained and optimized by setting model parameters, such as the number of decision trees and the maximum depth of each decision tree. During the model training process, a cross-validation method is used to evaluate model performance. The data set is divided into a training set and a validation set. The average performance indicators of the model, such as the root mean square error and the coefficient of determination, are obtained through multiple cross-validations. Based on the cross-validation results, a random forest model is used to calibrate and adjust the parameters, and the optimal parameter combination is found as the optimal parameters through methods such as grid search.

[0120] For example, cross-validation is an important means of evaluating model performance. Using 5-fold cross-validation, the data set is randomly divided into 5 parts, 4 parts are used for training each time, and 1 part is used for validation, and this is repeated 5 times. In this way, the performance of the model on different data subsets can be obtained, and the model performance can be evaluated more comprehensively. For example, in a certain validation, the root mean square error of the model on the training set was 0.3 mg / L, and on the validation set it was 0.4 mg / L, indicating that the model has good generalization ability. Parameter calibration is a key step in optimizing model performance. Through the grid search method, different parameter combinations can be systematically tried. For example, the number of decision trees ranges from 100 to 1000, with a step size of 100; the maximum depth ranges from 5 to 25, with a step size of 5. Each combination is cross-validated, and the parameters with the best performance on the validation set are selected.

[0121] In step S4, the prediction model is adjusted according to the optimization parameters to obtain the optimal prediction model.

[0122] In the above step S4, the prediction model is adjusted according to the optimization parameters to obtain the optimal prediction model, which specifically includes the following steps:

[0123] S41, setting the prediction model according to the optimization parameters to obtain an optimized model;

[0124] S42, performing prediction based on the optimization model to obtain a predicted value;

[0125] S43, subtracting the predicted value from the pre-stored measured value to obtain a deviation value;

[0126] S44, making a judgment based on the deviation value and the preset threshold range; if the deviation value is within the preset threshold range, the optimization model is determined to be the optimal prediction model; if the deviation value exceeds the preset threshold range, the optimization model is iteratively optimized using the neural network structure optimization technology to obtain a new model and repeat the above operation until the optimal prediction model is obtained.

[0127] In a specific embodiment, the above steps S41 to S44 are specifically implemented as follows: applying the calibrated optimization parameters of the random forest model to a new model to obtain an optimized model, inputting relevant multi-factor coupling information data, and the model predicts the dissolved oxygen concentration according to the decision rules obtained by training to obtain a prediction result; calculating the deviation value between the prediction result and the measured data; comparing the deviation value with a preset deviation threshold to determine whether the threshold is exceeded; if the deviation value exceeds the threshold, determining the model parameters that need to be adjusted based on the size and direction of the deviation; iteratively adjusting the model parameters through an optimization algorithm such as a gradient descent method to obtain a more accurate prediction result; Updated model parameters; according to the complexity of the model structure and the requirements of prediction accuracy, the neural network structure optimization technology, such as network pruning and layer adjustment, is used to optimize the model structure; through correlation analysis and feature selection algorithms, new influencing factors related to the prediction target are mined from the data, introduced into the model, and the feature space of the model is expanded; using the updated model parameters, the optimized model structure and the newly introduced influencing factors, the model is recalibrated to obtain a new model with higher accuracy, and the new model is recalculated and judged in steps S41 to S44, and it is continuously iterated until a model that meets the conditions is obtained and output as the optimal prediction model.

[0128] For example, analyzing the deviation between model predictions and measured data is a key step in evaluating model performance. By calculating the difference between predicted and measured values, model accuracy can be quantified. For example, for dissolved oxygen concentration prediction, if the model predicts 7.5 mg / L and the measured value is 8.0 mg / L, the deviation is 0.5 mg / L. This deviation value needs to be compared with a pre-set threshold to determine whether the model needs adjustment. The setting of the deviation threshold should take into account the actual application scenario and accuracy requirements. In water quality monitoring, the deviation threshold is set at 0.3 mg / L. If the calculated deviation exceeds this threshold, the model needs to be adjusted. The magnitude and direction of the deviation provide guidance for parameter adjustment. For example, if the predicted values are consistently lower than the measured values, it is necessary to increase the weights of certain positively correlated factors. Optimization algorithms such as gradient descent are commonly used for parameter adjustment. In a random forest model, parameters such as the number of trees and tree depth can be adjusted. Assuming an initial setting of 100 trees with a maximum depth of 10, gradient descent can optimize the model to 120 trees with a maximum depth of 12. This adjustment can increase the model's complexity, allowing it to capture more subtle data features. Optimizing model structure is another important means of improving prediction accuracy. For neural network models, network pruning techniques can be used to remove unimportant connections or adjust the number of network layers. For example, a five-layer fully connected network can be optimized to a three-layer structure, reducing the risk of overfitting and improving computational efficiency. Feature selection and the introduction of new influencing factors have a significant impact on model performance. Correlation analysis revealed a strong negative correlation between water temperature and dissolved oxygen concentration, while algae concentration is positively correlated with dissolved oxygen. These newly discovered influencing factors can be introduced into the model to expand the feature space. For example, adding water temperature and algae concentration as new features to existing factors such as pH and water depth can significantly improve the model's prediction accuracy. Model recalibration is the final step in the optimization process. Using the updated parameters, optimized structure, and newly introduced features, the model is fully retrained and tuned. This process requires multiple iterations, with model performance evaluated each time until the target accuracy is achieved.

[0129] In step S5, a layout analysis is performed based on the concentration data to obtain a monitoring layout plan.

[0130] In the above step S5, the layout analysis is performed based on the concentration data to obtain a monitoring layout plan, which specifically includes the following steps:

[0131] S51, performing position reading on the concentration data to obtain position information;

[0132] S52, integrating the concentration data and the location information to obtain monitoring point data;

[0133] S53, performing spatial difference analysis based on the monitoring point data to obtain a spatial distribution map;

[0134] S54, superimposing the spatial distribution map with a pre-stored water body function zoning layer, a pre-stored pollution source distribution layer, and pre-stored hydrological dynamic condition data to obtain a superposition result;

[0135] S55, identifying according to the superposition result to obtain spatial relationships and influencing factors;

[0136] S56, constructing a model based on the spatial relationship and the influencing factors to obtain a concentration distribution prediction model;

[0137] S57, based on the concentration distribution prediction model, optimize the spatial and temporal representativeness to obtain the monitoring layout plan.

[0138] It should be noted that the specific implementation process of steps S51 to S57 above includes: obtaining dissolved oxygen concentration monitoring data from multiple monitoring points within the study area, and simultaneously obtaining the geographic location information of each monitoring point; importing the geographic coordinates of the monitoring points and the dissolved oxygen concentration data into geographic information system software to construct a spatial database; selecting an appropriate spatial interpolation method, such as kriging interpolation or inverse distance weighted interpolation, to perform spatial interpolation analysis on the monitoring point data to generate a spatial distribution map of dissolved oxygen concentration in the study area; and overlaying these pre-stored layers of water function zoning, pollution source distribution, and hydrological and dynamic conditions data with the spatial distribution map of dissolved oxygen concentration. Based on the overlay analysis results, the spatial relationship and influencing factors between dissolved oxygen concentration distribution and water function zoning, pollution source distribution, and hydrological and dynamic conditions are identified. A machine learning algorithm, such as a decision tree or support vector machine, is used to establish a predictive model between dissolved oxygen concentration distribution and influencing factors, and the spatial and temporal representativeness of different monitoring point layout plans is evaluated. Based on the evaluation results of the predictive model, the monitoring point layout plan is optimized.

[0139] For example, 50 monitoring points were set up in the Taihu Lake basin. Dissolved oxygen data were collected four times daily at each location, and their latitude and longitude coordinates were recorded. This raw data formed the basis for spatial analysis. Importing this data into geographic information system software such as ArcGIS can visually display the spatial distribution of the monitoring points. However, discrete point data alone cannot accurately reflect the dissolved oxygen distribution across the entire region. Therefore, spatial interpolation methods are needed to generate a continuous dissolved oxygen distribution map. Kriging interpolation accounts for spatial autocorrelation and is suitable for interpolating environmental factors. This method can generate estimated dissolved oxygen values for any location within the study area, generating raster data. The generation of spatial distribution maps can identify areas of high and low dissolved oxygen values. For example, dissolved oxygen concentrations are generally low along the northern shore of Taihu Lake, while higher in the center of the lake. This distribution pattern is influenced by complex factors. For in-depth analysis, more spatial information is needed. The water function zoning layer reflects the uses of different areas, such as drinking water source protection areas and fishery areas. The pollution source distribution layer includes information on industrial outfalls and agricultural non-point source pollution areas. Hydrodynamic data includes elements such as flow direction and velocity. Overlaying these layers with the dissolved oxygen distribution map reveals some interesting correlations. For example, dissolved oxygen concentrations are significantly lower in industrially dense bays than in surrounding areas, while dissolved oxygen levels are relatively high at river junctions with faster currents. These spatial relationships provide a basis for developing a predictive model. Using a decision tree algorithm, a highly interpretable model can be constructed. For example, the model reveals that the probability of a drop in dissolved oxygen concentration increases significantly when the distance to the pollution source is less than 500 meters and the water velocity is less than 0.1 m / s. This model not only predicts dissolved oxygen distribution but also assesses the representativeness of monitoring points. Based on the model evaluation results, the monitoring network can be optimized. For example, if five of the original 50 points were found to have high information redundancy, they could be considered for removal. At the same time, three new monitoring points were added in areas sensitive to dissolved oxygen fluctuations. This optimization ensures data representativeness while improving monitoring efficiency.

[0140] In step S6, a threshold analysis is performed based on the water body function, the environmental capacity and the ecological target to obtain a threshold solution.

[0141] In the above step S6, the threshold analysis is performed based on the water body function, the environmental capacity and the ecological target to obtain a threshold solution, which specifically includes the following steps:

[0142] S61, building a model based on the water body function, the environmental capacity, and the ecological goal to obtain a threshold evaluation index;

[0143] S62, performing index calculation according to the threshold evaluation index to obtain a comprehensive evaluation value;

[0144] S63: Generate a plan using an algorithm based on the comprehensive evaluation value and pre-stored management requirements to obtain a threshold plan.

[0145] In one implementation, steps S61 to S63 specifically include: establishing an evaluation index system and model for dissolved oxygen concentration thresholds using a fuzzy comprehensive evaluation method based on relevant attributes such as water body function, environmental capacity, and ecological objectives; determining the weights of each evaluation index using the analytic hierarchy process (AHP) to calculate a comprehensive evaluation value for the dissolved oxygen concentration threshold; obtaining opinions and suggestions from different departments on dissolved oxygen concentration threshold management in response to multi-departmental management needs; performing text mining and sentiment analysis using natural language processing techniques to extract keywords and sentiment trends, and determining the level of support for the threshold management plan from different departments; and obtaining public comment data on water environment quality on social media platforms using web crawling technology in response to public participation. Classifying the comment data using machine learning algorithms such as support vector machines or naive Bayesian algorithms to identify public satisfaction and concerns about water environment quality. Based on the comprehensive evaluation value of the dissolved oxygen concentration threshold, combined with support for multi-departmental management needs and public participation satisfaction, a decision tree algorithm is used to automatically generate a warning level classification scheme for the dissolved oxygen concentration threshold, determine the dissolved oxygen concentration threshold ranges corresponding to different warning levels, and generate a threshold scheme.

[0146] It should be noted that the threshold scheme refers to the threshold management scheme, which specifically includes clear warning levels, response measures and information disclosure mechanisms. The implementation process of the threshold management scheme for different aspects is as follows: For the warning level attribute, rule engine technology is used to automatically determine the warning level of the current dissolved oxygen concentration based on the preset IF-THEN rules. If the dissolved oxygen concentration is lower than a certain warning level threshold, the corresponding warning information is triggered. For the response measure attribute, case reasoning technology is used to retrieve historical cases similar to the current water body situation based on water body functions, environmental capacity and other attributes, obtain the response measures taken in the case, and make appropriate adjustments based on the actual situation to form a targeted response measure plan; for the information disclosure attribute, according to the information disclosure mechanism, a public template for information such as the dissolved oxygen concentration warning level and response measures is automatically generated. Through data visualization technology, the relevant information is intuitively displayed in the form of charts, maps, etc., and released to the public through channels such as websites and mobile apps to improve the timeliness and transparency of information disclosure.

[0147] In step S7, the optimal prediction model and the threshold solution are integrated to obtain an early warning solution.

[0148] For example, a dissolved oxygen concentration prediction model and threshold management plan are input into the system to build a real-time early warning system. Specifically, early warning information can be released through multiple channels, including websites and mobile devices, using graphics, text, and audio to present the warning content.

[0149] In step S8, the monitoring layout plan and the optimization parameters are adjusted according to the feedback of the early warning plan in the system to obtain an early warning optimization plan.

[0150] In one feasible approach, based on the operational feedback data of the early warning system, a machine learning algorithm is used to optimize the monitoring point layout plan, and the optimal monitoring point distribution is determined through cluster analysis; the historical prediction results and actual water quality data of the early warning model are obtained, and the model parameters are dynamically adjusted using the gradient descent algorithm to minimize the prediction error; the water quality monitoring data are analyzed using an association rule mining algorithm to discover the correlation patterns between different water quality indicators, providing new features for water quality early warning; the water quality data are trend-forecasted using a time series analysis method to determine the changing trend of water quality in the future and improve the timeliness of the early warning; the optimized monitoring point layout plan and model parameters are applied to the early warning system to improve the accuracy and reliability of water quality early warning; based on the water quality early warning results, a decision tree algorithm is used to generate decision rules for water quality improvement and ecological restoration, providing management departments with feasible action plans; the operational feedback data of the early warning system is continuously collected, and the monitoring point layout, model parameters and data mining processes are regularly optimized and iterated to obtain an early warning optimization plan and form a closed-loop early warning optimization mechanism.

[0151] In step S9, the system is configured according to the early warning optimization scheme so that the system operates according to the early warning optimization scheme.

[0152] Exemplarily, the important parameters in the early warning optimization scheme are input into the system or the system is directly connected to operate via data transmission.

[0153] In summary, the present invention discloses a real-time monitoring and early warning method for water environment based on the Internet of Things, including obtaining concentration data and water body attributes; wherein the water body attributes include water body function, environmental capacity and ecological goals; performing feature analysis based on the concentration data to obtain a prediction model; performing optimization based on the prediction model to obtain optimization parameters; adjusting the prediction model based on the optimization parameters to obtain an optimal prediction model; performing layout analysis based on the concentration data to obtain a monitoring layout plan; performing threshold analysis based on the water body function, the environmental capacity and the ecological goals to obtain a threshold plan; integrating the optimal prediction model and the threshold plan to obtain an early warning plan; adjusting the monitoring layout plan and the optimization parameters based on the feedback of the early warning plan in the system to obtain an early warning optimization plan; setting the system according to the early warning optimization plan so that the system works according to the early warning optimization plan. This method uses a sensor network to collect high-frequency, multi-point dissolved oxygen concentration data. It then integrates factors such as temperature, water velocity, and eutrophication to construct a dissolved oxygen concentration prediction model. Time series analysis and a random forest algorithm are used to process the data and calibrate model parameters. Spatial interpolation is performed using a geographic information system to generate a dissolved oxygen concentration distribution map. Based on water body functions and ecological objectives, dissolved oxygen concentration thresholds are determined to construct a real-time early warning system. By dynamically adjusting monitoring schemes and model parameters, this method continuously optimizes early warning accuracy and timeliness, providing decision-making support for water quality improvement and ecological restoration. This enables precise monitoring, prediction, and early warning of dissolved oxygen concentration in water, enhancing the scientific nature and effectiveness of water environment management.

[0154] Reference Figure 2 The second embodiment of the present invention provides a water environment real-time monitoring and early warning system based on the Internet of Things, including:

[0155] Data acquisition module 101, used to acquire concentration data and water body attributes; wherein the water body attributes include water body function, environmental capacity and ecological objectives;

[0156] A model building module 102 is used to perform feature analysis based on the concentration data to obtain a prediction model;

[0157] A parameter optimization module 103 is used to perform optimization according to the prediction model to obtain optimized parameters;

[0158] A model optimization module 104 is configured to adjust the prediction model according to the optimization parameters to obtain an optimal prediction model;

[0159] A layout analysis module 105 is configured to perform layout analysis based on the concentration data to obtain a monitoring layout plan;

[0160] A threshold analysis module 106 is configured to perform a threshold analysis based on the water body function, the environmental capacity, and the ecological target to obtain a threshold solution;

[0161] Integration module 107, used to integrate the optimal prediction model and the threshold solution to obtain an early warning solution;

[0162] Feedback adjustment module 108, configured to adjust the monitoring deployment plan and the optimization parameters according to the feedback of the early warning plan in the system to obtain an early warning optimization plan;

[0163] The system setting module 109 is used to set the system according to the early warning optimization plan so that the system works according to the early warning optimization plan.

[0164] It should be noted that the real-time monitoring and early warning system for water environment based on the Internet of Things provided in an embodiment of the present invention is used to execute all the process steps of the real-time monitoring and early warning method for water environment based on the Internet of Things in the above embodiment. The working principles and beneficial effects of the two correspond one to one, so they will not be repeated here.

[0165] An embodiment of the present invention further provides an electronic device. The electronic device includes: a processor, a memory, and a computer program stored in the memory and executable on the processor, such as a program for a method for real-time monitoring and early warning of a water environment based on the Internet of Things. When the processor executes the computer program, the steps in each of the above-mentioned embodiments of the method for real-time monitoring and early warning of a water environment based on the Internet of Things are implemented, such as Figure 1 Alternatively, when the processor executes the computer program, the functions of the modules / units in the above-mentioned device embodiments are realized, such as the data acquisition module.

[0166] Exemplarily, the computer program may be divided into one or more modules / units, which are stored in the memory and executed by the processor to implement the present invention. The one or more modules / units may be a series of computer program instruction segments capable of implementing specific functions, and the instruction segments are used to describe the execution process of the computer program in the electronic device.

[0167] The electronic device may be a computing device such as a desktop computer, notebook, PDA, or smart tablet. The electronic device may include, but is not limited to, a processor and memory. Those skilled in the art will appreciate that the aforementioned components are merely examples of electronic devices and do not constitute a limitation of the electronic device. The electronic device may include more or fewer components than those described above, or a combination of certain components, or different components. For example, the electronic device may also include input / output devices, network access devices, buses, and the like.

[0168] The processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the electronic device, connecting various parts of the entire electronic device using various interfaces and lines.

[0169] The memory can be used to store the computer programs and / or modules, and the processor realizes various functions of the electronic device by running or executing the computer programs and / or modules stored in the memory and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, at least one application required for a function (such as a sound playback function, an image playback function, etc.); the data storage area can store data created based on the use of the mobile phone (such as audio data, a phone book, etc.). In addition, the memory can include a high-speed random access memory and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage device.

[0170] Wherein, if the module / unit integrated in the electronic device is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention implements all or part of the process in the above-mentioned embodiment method, and can also be completed by a computer program to instruct the relevant hardware. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by the processor, it can implement the steps of each of the above-mentioned method embodiments. Wherein, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.

[0171] It should be noted that the device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed across multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the present embodiment. In addition, in the drawings of the device embodiments provided by the present invention, the connection relationship between the modules indicates that there is a communication connection between them, which may be specifically implemented as one or more communication buses or signal lines. A person of ordinary skill in the art can understand and implement the present invention without inventive effort.

[0172] The specific embodiments described above further illustrate the objectives, technical solutions, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. In particular, it should be noted that any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included within the scope of protection of the present invention for those skilled in the art.

Claims

1. A real-time monitoring and early warning method for water environment based on the Internet of Things, characterized in that: Executed by a computer, including: Acquiring concentration data and water body attributes; wherein the water body attributes include water body function, environmental capacity, and ecological objectives; Performing feature analysis based on the concentration data to obtain a prediction model; Optimize according to the prediction model to obtain optimized parameters; Adjusting the prediction model according to the optimization parameters to obtain an optimal prediction model; Performing layout analysis based on the concentration data to obtain a monitoring layout plan; Performing a threshold analysis based on the water body function, the environmental capacity, and the ecological goal to obtain a threshold solution; Integrating the optimal prediction model and the threshold scheme to obtain an early warning scheme; Adjust the monitoring layout plan and the optimization parameters according to the feedback of the early warning plan in the system to obtain an early warning optimization plan; The system is configured according to the early warning optimization plan so that the system works according to the early warning optimization plan.

2. The water environment real-time monitoring and early warning method based on the Internet of Things according to claim 1 is characterized in that: The method of performing feature analysis based on the concentration data to obtain a prediction model includes: Decomposing the concentration data to obtain trend item data, period item data and random item data; Perform long-term fitting based on the trend item data to obtain trend change rules; Transform and extract the periodic item data to obtain a periodic variation pattern; Perform short-term fitting based on the random item data to obtain random fluctuation patterns; A prediction model is constructed based on the trend change law, the period change law and the random fluctuation law.

3. The water environment real-time monitoring and early warning method based on the Internet of Things according to claim 1 is characterized in that: The step of evaluating the prediction model to obtain optimized parameters includes: The prediction model is evaluated using a cross-validation method to obtain a cross-validation result; Calibration is performed based on the cross-validation results to obtain optimized parameters.

4. The water environment real-time monitoring and early warning method based on the Internet of Things according to claim 1 is characterized in that: The step of adjusting the prediction model according to the optimization parameters to obtain an optimal prediction model includes: Setting the prediction model according to the optimization parameters to obtain an optimized model; Perform prediction according to the optimization model to obtain a predicted value; Subtracting the predicted value from the pre-stored measured value to obtain a deviation value; A judgment is made based on the deviation value and the preset threshold range; if the deviation value is within the preset threshold range, the optimization model is determined to be the optimal prediction model; if the deviation value exceeds the preset threshold range, the optimization model is iteratively optimized using the neural network structure optimization technology to obtain a new model and the above operation is repeated until the optimal prediction model is obtained.

5. The water environment real-time monitoring and early warning method based on the Internet of Things according to claim 4 is characterized in that: The method of iteratively optimizing the optimization model using a neural network structure optimization technique to obtain a new model includes: Performing structural optimization according to the optimization model to obtain a new model structure; Mining is performed based on the relevance and characteristics of the optimization model to obtain influencing factors; The optimized model is recalibrated according to the new model structure and the influencing factors to obtain a new model.

6. The method for real-time monitoring and early warning of water environment based on Internet of Things according to claim 1, characterized in that: The performing layout analysis based on the concentration data to obtain a monitoring layout plan includes: Performing position reading on the concentration data to obtain position information; Integrate the concentration data and location information to obtain monitoring point data; Performing spatial difference analysis based on the monitoring point data to obtain a spatial distribution map; Superimposing the spatial distribution map with a pre-stored water body function zoning layer, a pre-stored pollution source distribution layer, and pre-stored hydrological dynamic condition data to obtain a superposition result; Identify based on the superposition results to obtain spatial relationships and influencing factors; Constructing a model based on the spatial relationship and the influencing factors to obtain a concentration distribution prediction model; The temporal and spatial representativeness is optimized based on the concentration distribution prediction model to obtain the monitoring layout plan.

7. The water environment real-time monitoring and early warning method based on the Internet of Things according to claim 1 is characterized in that: The threshold analysis is performed based on the water body function, the environmental capacity and the ecological target to obtain a threshold solution, including: Building a model based on the water body function, the environmental capacity and the ecological goal to obtain a threshold evaluation index; Performing index calculation according to the threshold evaluation index to obtain a comprehensive evaluation value; An algorithm is used to generate a solution based on the comprehensive evaluation value and pre-stored management requirements to obtain a threshold solution.

8. The method for real-time monitoring and early warning of water environment based on Internet of Things according to claim 7 is characterized in that: The threshold scheme refers to a scheme that can clearly define the warning level, response measures and information disclosure mechanism; for the warning level attribute, the rule engine technology can be used to automatically determine the warning level of the current dissolved oxygen concentration based on the preset IF-THEN rules. If the dissolved oxygen concentration is lower than a certain warning level threshold, the corresponding warning information is triggered; for the response measure attribute, the case reasoning technology is used to retrieve historical cases similar to the current water body situation based on the water body function and environmental capacity attributes, obtain the response measures taken in the case, and make appropriate adjustments based on the actual situation to form a targeted response measure plan; for the information disclosure attribute, according to the content of the plan, a public template for information such as the dissolved oxygen concentration warning level and response measures is automatically generated for display.

9. A real-time water environment monitoring and early warning system based on the Internet of Things, characterized in that: include: A data acquisition module, configured to acquire concentration data and water attributes, wherein the water attributes include water function, environmental capacity, and ecological objectives; A model building module, used to perform feature analysis based on the concentration data to obtain a prediction model; A parameter optimization module, configured to perform optimization based on the prediction model to obtain optimized parameters; A model optimization module, configured to adjust the prediction model according to the optimization parameters to obtain an optimal prediction model; A layout analysis module, configured to perform layout analysis based on the concentration data to obtain a monitoring layout plan; A threshold analysis module, configured to perform a threshold analysis based on the water body function, the environmental capacity, and the ecological target to obtain a threshold solution; An integration module, configured to integrate the optimal prediction model and the threshold solution to obtain an early warning solution; A feedback adjustment module, configured to adjust the monitoring deployment plan and the optimization parameters according to the feedback of the early warning plan in the system to obtain an early warning optimization plan; The system setting module is used to set the system according to the early warning optimization plan so that the system works according to the early warning optimization plan.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium includes a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute the real-time water environment monitoring and early warning method based on the Internet of Things as described in any one of claims 1 to 8.

Images