Environmental risk comprehensive prevention and control platform

Abstract

The invention provides an environmental risk comprehensive prevention and control platform, which relates to the technical field of environmental risk prevention and control and comprises a data acquisition module, a data preprocessing module, a multi-index association analysis module, a risk early warning module and a user interface module. Wherein the multi-index correlation analysis module is used for mining an influence rule between a frequent item set of a ph value, a water temperature and a heavy metal concentration and a water body transmissivity through big data, and mining an influence rule between a time data sequence of meteorological data and a nutritive salt concentration and a chlorophyll concentration, so that relatively accurate water body transmissivity and chlorophyll concentration can be obtained through analysis; therefore, a relatively accurate environmental risk event can be predicted. Through the user interface module, a result obtained through big data analysis can be displayed in an interface in a man-machine interaction mode, and therefore a manager can conveniently conduct access and processing at any time.

Description

Technical Field

[0001] The present invention relates to the technical field of environmental risk prevention and control, and in particular to an environmental risk comprehensive prevention and control platform. Background Art

[0002] Water resource environmental risk prevention and control requires comprehensive and dynamic monitoring of water quality. Traditional methods usually rely on a single or a small number of indicators (such as pH value, dissolved oxygen, etc.), which are difficult to fully reflect the complexity of the water environment and the diversity of pollutants. For example, changes in the concentration of certain pollutants (such as heavy metals and organic pollutants) may have serious impacts on the aquatic ecosystem and human health, while the optical properties of water bodies (such as transmittance and turbidity) can indirectly reflect suspended matter, algae reproduction, etc. However, there is currently no systematic method to combine these multi-dimensional indicators to achieve accurate risk prediction and prevention and control through big data analysis and machine learning technology. Summary of the invention

[0003] The purpose of the present invention is to provide an integrated environmental risk prevention and control platform to solve the technical problem that the prior art cannot accurately predict environmental risks.

[0004] This application proposes an environmental risk comprehensive prevention and control platform, including: Data collection module: used to collect the first multi-dimensional monitoring data related to water resources environment in real time; Data preprocessing module: cleaning, normalizing and feature extracting the collected first multi-dimensional monitoring data to obtain second multi-dimensional monitoring data; Multi-index correlation analysis module: performing correlation analysis on the second multi-dimensional monitoring data to determine the first water body transmittance and the first chlorophyll concentration; The first chlorophyll concentration is determined by the following steps: S21: Acquire a first meteorological change sequence of first meteorological data within a preset time interval before a first time point; S22: determining a first nutrient concentration according to nutrient concentrations of multiple types of water bodies at a first time point; S23: inputting the first meteorological change sequence and the first nutrient salt concentration into a first chlorophyll prediction model to obtain a first chlorophyll concentration; Risk warning module: determining a first risk probability and issuing a risk warning according to the first water body transmittance; User interface module: real-time display of the first multi-dimensional monitoring data and the first water body transmittance.

[0005] Preferably, the first multi-dimensional monitoring data includes: heavy metal concentration, nutrient salt concentration, pH value, water temperature, and meteorological data.

[0006] Preferably, the first water body transmittance is determined by the following steps: S11: discretize the first original information combination to obtain a first input information combination; S12: Inputting the first input information combination into a first water body transmittance prediction model to obtain a first water body transmittance.

[0007] Preferably, the first original information combination includes: A first water temperature, a first pH value, and a first heavy metal concentration.

[0008] Preferably, the S11 includes: Acquire multiple groups of first sample data; Clustering multiple groups of the first sample data to obtain multiple first cluster centers; Determine a plurality of first partition values ​​according to a plurality of the first cluster centers; The first original information combination is discretized according to the plurality of the first division values ​​to obtain a first input information combination.

[0009] Preferably, the multiple types of water bodies include: Surface water, stratified water and mixed water. The surface water refers to the water 0.5 meters below the water surface, the stratified water refers to the collection of water at different depths, and the mixed water refers to the mixed water from multiple sampling points.

[0010] Preferably, the first chlorophyll prediction model comprises: Input layer: Input data includes the nutrient concentration at a specified time point and the meteorological change sequence within a preset time interval before the specified time point; LSTM layer: used to analyze the long-term dependencies of the data in the meteorological change sequence; Fully connected layer: outputs the features extracted from the LSTM layer and the nutrient concentration at the specified time point; Output layer: Output the first chlorophyll concentration.

[0011] The application proposes a comprehensive environmental risk prevention and control platform, which relates to the field of environmental risk prevention and control technology, including a data acquisition module, a data preprocessing module, a multi-index correlation analysis module, a risk warning module and a user interface module, wherein the multi-index correlation analysis module mines the influence rules between the frequent item sets of pH value, water temperature, heavy metal concentration and water body transmittance through big data, and mines the influence rules between the time data series of meteorological data, nutrient concentration and chlorophyll concentration, and can analyze and obtain more accurate water body transmittance and chlorophyll concentration, so as to predict more accurate environmental risk events. Through the user interface module, the results of big data analysis can be displayed in the interface through human-computer interaction, so that managers can access and process them at any time. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] In order to more clearly illustrate the implementation methods of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for the implementation methods or the description of the prior art. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other implementation drawings can be derived from the provided drawings without creative work.

[0013] Figure 1 It is a system structure diagram of an environmental risk comprehensive prevention and control platform in the present invention.

[0014] Figure 2 This is a flow chart for determining the transmittance of the first water body in the present invention. DETAILED DESCRIPTION

[0015] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0016] The present invention will be described in detail below in conjunction with the accompanying drawings and specific embodiments, wherein the illustrative embodiments and descriptions are only used to explain the present invention but are not intended to limit the present invention.

[0017] The following is a detailed description of an environmental risk comprehensive prevention and control platform of the present invention.

[0018] This embodiment proposes an environmental risk comprehensive prevention and control platform, the system structure of the platform is as follows Figure 1 As shown, it includes a data acquisition module, a data preprocessing module, a multi-indicator correlation analysis module, a risk warning module and a user interface module.

[0019] Data acquisition module: used to collect multi-dimensional monitoring data related to water resources and environment in real time, including but not limited to: pollutant concentration indicators: such as the concentration of heavy metals (lead, mercury, cadmium, etc.), organic pollutants (benzene, polycyclic aromatic hydrocarbons, etc.), and nutrients (ammonia nitrogen, total phosphorus, etc.); optical property indicators: such as water body transmittance, etc.; conventional water quality indicators: such as pH value, dissolved oxygen, conductivity, temperature, etc.; meteorological and hydrological data: such as rainfall, wind speed, water flow speed, etc.

[0020] Specifically, the data acquisition module collects multi-dimensional water environment data in real time, including pollutant concentrations, optical properties, conventional water quality indicators, and meteorological and hydrological data, through a sensor network deployed in various water sources.

[0021] Data preprocessing module: cleans, normalizes and extracts features from the collected raw data to improve data quality and prepare it for subsequent analysis.

[0022] Specifically, the data preprocessing module cleans the collected data, removes noise and outliers, then performs normalization, and finally extracts useful features.

[0023] Multi-indicator correlation analysis module: Use big data and machine learning models to conduct correlation analysis on multi-dimensional indicators, explore the intrinsic connection between indicators such as pollutant concentration and optical properties, and their correlation with water environment pollution risks.

[0024] The multi-index correlation analysis module is used to analyze the influence relationship between heavy metal concentration and water body transmittance, and between nutrient concentration and chlorophyll.

[0025] (1) Analysis of the influence relationship between heavy metal concentration and water transmittance In addition to heavy metal concentration, water transmittance is also affected by factors such as water temperature and pH value. In this step, the relevant historical data of the above factors are discretized and frequent item sets with specific relationships with water transmittance are formed, thereby providing sample data for the big data prediction model.

[0026] The method steps for determining the influence relationship between heavy metal concentration and water transmittance are as follows: Figure 2 As shown: S11: discretize the first original information combination to obtain a first input information combination.

[0027] The first original information combination includes the first water temperature, the first pH value, the first chlorophyll a concentration and the first heavy metal concentration received in real time. In order to be able to combine the above indicators to analyze the change law between the water body transmittance, the above information needs to be discretized in this step.

[0028] Preferably, the above discretization operation can be performed with the help of expert experience. For example, when discretizing the water temperature, low temperature: <10°C, medium temperature: 10°C-20°C, high temperature: >20°C, through the above discretization process, a discretization index for characterizing the first water temperature can be determined, that is, low temperature, medium temperature or high temperature.

[0029] Preferably, the discretized first division point value can also be obtained through big data clustering, and the above discretization operation can be performed according to the first division point value. Specifically, multiple groups of first water temperature sample data are first obtained, and multiple groups of the first water temperature sample data are clustered to obtain multiple first cluster centers, multiple first division values ​​are determined according to the multiple first cluster centers, and the corresponding indicators are discretized according to the multiple first division values. For example, when discretizing the water temperature, multiple groups of water temperature sample data are first obtained, and multiple first cluster centers are obtained after clustering. Assuming that the multiple cluster centers are 10°C, 18°C ​​and 22°C respectively, 10°C, 18°C ​​and 22°C can be used as the three first division point values, respectively, so as to determine multiple discrete indicators for characterizing the first water temperature between the first division point values.

[0030] S12: Inputting the first input information combination into a first water body transmittance prediction model to obtain a first water body transmittance.

[0031] In the S11, the indicators such as water temperature, pH value, turbidity and heavy metal concentration are discretized respectively, and the first input information combination is formed based on the discretized indicators. In order to predict the water body transmittance according to the first input information combination, the first input information combination is input into the pre-trained first water body transmittance prediction model to obtain the first water body transmittance.

[0032] The first water body transmittance prediction model is obtained by training in the following manner: First, sample data is obtained. For water bodies with serious water pollution, the discrete indicators of water temperature, pH value and heavy metal concentration are obtained, and the water body transmittance is obtained. Sample data of several water bodies are obtained according to the above method.

[0033] Secondly, a convolutional neural network model is trained to obtain the first water body transmittance prediction model. The information combination consisting of the water temperature, pH value and heavy metal concentration is used as input data, and the water body transmittance is used as output data to train the convolutional neural network model to obtain the first water body transmittance prediction model. During the prediction process, the first water body transmittance prediction model mainly determines the most relevant water body transmittance by calculating the Euclidean distance between different input information combinations.

[0034] (2) Combined with meteorological data (such as rainfall), the relationship between nutrient concentration and chlorophyll concentration is analyzed to predict the risk of algae blooms. Chlorophyll concentration is an important factor in predicting the risk of algae outbreaks, and it is closely related to meteorological data such as rainfall and nutrient concentration data in water bodies. However, the impact of meteorological data such as rainfall on water bodies often has a certain lag. For example, short-term high-intensity rainfall and long-term low-intensity rainfall often have different effects on algae outbreaks in water bodies.

[0035] The steps for determining chlorophyll concentration are as follows: S21: Acquire a first meteorological change sequence of first meteorological data within a preset time interval before a first time point.

[0036] Preferably, the first meteorological data is rainfall data, and the preset time interval is preferably 6 hours.

[0037] The first meteorological change sequence is composed of a plurality of first numerical points. Taking the first meteorological data as rainfall data as an example, each of the first numerical points includes the cumulative rainfall duration up to the first moment, and the average rainfall intensity within the cumulative rainfall duration. That is, each of the first numerical points in the first meteorological change sequence includes a one-dimensional vector [cumulative rainfall duration, average rainfall intensity within the cumulative rainfall duration].

[0038] S22: Determine a first nutrient concentration according to nutrient concentrations of multiple types of water bodies at a first time point.

[0039] The multiple types of water bodies include surface water, stratified water and mixed water. The surface water sample refers to the water sample at 0.5 meters below the water surface, the stratified water sample refers to the collection of water samples at different water depths (0.5 meters, 1 meter, 1.5 meters, etc.), and the mixed water refers to the mixture of water samples from multiple sampling points.

[0040] In this step, a designated operation is performed on the nutrient salt concentration in the water sample obtained by the above three methods to determine the first nutrient salt concentration.

[0041] The specified operation is preferably a weighted summation.

[0042] S23: Inputting the first meteorological change sequence and the first nutrient salt concentration into a first chlorophyll prediction model to obtain a first chlorophyll concentration.

[0043] The first chlorophyll prediction model includes: Input layer: The input data includes the nutrient concentration at a specified time point and the meteorological change sequence within a preset time interval before the specified time point.

[0044] LSTM layer: used to analyze the long-term dependencies of the data in the meteorological change sequence.

[0045] Fully connected layer: outputs the features extracted in the LSTM layer and the nutrient concentration at the specified time point.

[0046] Output layer: Output chlorophyll concentration.

[0047] Among them, the specific structures of the LSTM layer and the fully connected layer, as well as the specific training method of the model can refer to the relevant technologies in the prior art.

[0048] Risk warning module: determines the first risk probability and issues a risk warning according to the first water body transmittance and the first chlorophyll concentration.

[0049] The first risk probability is positively correlated with the first water body transmittance and the first chlorophyll concentration.

[0050] User interface module: provides managers with intuitive data display and interactive interface to facilitate monitoring and decision-making.

[0051] The user interface module includes the following functions: (1) Map integration: For water bodies distributed by geographical location, the transmittance and chlorophyll concentration of different areas can be displayed on the map with different colors or marks to help managers understand geographical differences and hot spots.

[0052] (2) The system can receive and process data from sensors in real time and update it to the user interface in a timely manner, allowing managers to obtain the latest information.

[0053] (3) Provide tools such as filters and sliders so that users can customize the data range they view as needed (such as a specific time period, geographic location, etc.). Support clicking a point or area in the chart to pop up a detailed information box to display more detailed measurement data and other related information.

[0054] (4) Multi-platform compatibility. Ensure that the user interface can run smoothly on multiple devices such as desktop and mobile terminals to meet the needs of managers to access it anytime and anywhere.

[0055] The application proposes a comprehensive environmental risk prevention and control platform, which relates to the field of environmental risk prevention and control technology, including a data acquisition module, a data preprocessing module, a multi-index correlation analysis module, a risk warning module and a user interface module, wherein the multi-index correlation analysis module mines the influence rules between the frequent item sets of pH value, water temperature, heavy metal concentration and water body transmittance through big data, and mines the influence rules between the time data series of meteorological data, nutrient concentration and chlorophyll concentration, and can analyze and obtain more accurate water body transmittance and chlorophyll concentration, so as to predict more accurate environmental risk events. Through the user interface module, the results of big data analysis can be displayed in the interface through human-computer interaction, so that managers can access and process them at any time.

[0056] The above description is only a preferred embodiment of the present invention, so all equivalent changes or modifications made according to the structure, characteristics and principles described in the scope of the patent application of the present invention are included in the scope of the patent application of the present invention.

Claims

1. An environmental risk comprehensive prevention and control platform, characterized in that: include: Data collection module: used to collect the first multi-dimensional monitoring data related to water resources environment in real time; Data preprocessing module: cleaning, normalizing and feature extracting the collected first multi-dimensional monitoring data to obtain second multi-dimensional monitoring data; Multi-index correlation analysis module: performing correlation analysis on the second multi-dimensional monitoring data to determine the first water body transmittance and the first chlorophyll concentration; The first chlorophyll concentration is determined by the following steps: S21: Acquire a first meteorological change sequence of first meteorological data within a preset time interval before a first time point; S22: determining a first nutrient concentration according to nutrient concentrations of multiple types of water bodies at a first time point; S23: inputting the first meteorological change sequence and the first nutrient salt concentration into a first chlorophyll prediction model to obtain a first chlorophyll concentration; Risk warning module: determining a first risk probability and issuing a risk warning according to the first water body transmittance; User interface module: real-time display of the first multi-dimensional monitoring data and the first water body transmittance.

2. According to claim 1, a comprehensive environmental risk prevention and control platform is characterized in that: The first multi-dimensional monitoring data includes: heavy metal concentration, nutrient salt concentration, pH value, water temperature, and meteorological data.

3. The environmental risk comprehensive prevention and control platform according to claim 2 is characterized in that: The first water body transmittance is determined by the following steps: S11: discretize the first original information combination to obtain a first input information combination; S12: Inputting the first input information combination into a first water body transmittance prediction model to obtain a first water body transmittance.

4. The environmental risk comprehensive prevention and control platform according to claim 3 is characterized in that: The first original information combination includes: A first water temperature, a first pH value, and a first heavy metal concentration.

5. The environmental risk comprehensive prevention and control platform according to claim 4 is characterized in that: The S11 includes: Acquire multiple groups of first sample data; Clustering multiple groups of the first sample data to obtain multiple first cluster centers; Determine a plurality of first partition values ​​according to a plurality of the first cluster centers; The first original information combination is discretized according to the plurality of the first division values ​​to obtain a first input information combination.

6. The environmental risk comprehensive prevention and control platform according to claim 5 is characterized in that: The various types of water bodies include: Surface water, stratified water and mixed water. The surface water refers to the water 0.5 meters below the water surface, the stratified water refers to the collection of water at different depths, and the mixed water refers to the mixed water from multiple sampling points.

7. The environmental risk comprehensive prevention and control platform according to claim 6 is characterized in that: The first chlorophyll prediction model includes: Input layer: Input data includes the nutrient concentration at a specified time point and the meteorological change sequence within a preset time interval before the specified time point; LSTM layer: used to analyze the long-term dependencies of the data in the meteorological change sequence; Fully connected layer: outputs the features extracted from the LSTM layer and the nutrient concentration at the specified time point; Output layer: Output the first chlorophyll concentration.

Images