Environmental monitoring data visualization display platform and method

Abstract

The invention discloses an environmental monitoring data visualization display platform and method, and specifically relates to the technical field of data visualization. The invention divides a monitoring area into sub-monitoring areas, collects environmental data in each time period in real time, extracts transmission delay characteristics and data loss characteristics of the environmental data, calculates the real-time index of environmental data transmission, and divides the data into real-time, medium real-time and non-real-time data by comparing with a gradient standard threshold. The real-time data is used to generate visualization charts and maps, and the non-real-time data is marked as abnormal in a visualization interface. The medium real-time data is further analyzed for its real-time performance, and its use scope is adjusted and displayed by region, thereby solving the problem of data loss caused by satellite communication interruption in remote areas, ensuring that the environmental monitoring system can be dynamically updated and provide accurate decision support, and avoiding environmental events not being captured and reported in time.

Description

Technical Field

[0001] The present invention relates to the technical field of data visualization, and in particular to an environmental monitoring data visualization display platform and method. Background Art

[0002] Environmental monitoring data visualization refers to presenting complex environmental monitoring data in a graphical and visual way for more intuitive understanding and analysis. Environmental monitoring data usually includes information such as air quality, water quality, temperature, humidity, noise, and pollutant concentration. These data may be difficult to interpret in their original form, but through visualization methods such as charts, maps, heat maps, and trend charts, data changes, trends, and distribution can be more clearly displayed, helping users quickly obtain useful information.

[0003] In the prior art, when visualizing environmental monitoring data, first, sensors or monitoring equipment are used to collect environmental data, such as air quality, temperature, humidity, noise, etc. Secondly, after the data is collected, it is transmitted to the central server through the communication network for preprocessing, including data cleaning, outlier processing and formatting. Next, the preprocessed data is classified, aggregated or trend analyzed using data analysis algorithms. Then, the analysis results are graphically presented through visualization tools, such as line charts, bar charts, heat maps, geographic information system (GIS) maps, etc., to show the spatiotemporal distribution, trends and anomalies of the data. Finally, users view these visualization results in real time through the front-end interface (such as a web page or mobile application), and perform interactive operations as needed, such as zooming in, zooming out, and filtering different data levels.

[0004] The prior art has the following deficiencies:

[0005] In remote areas, satellite communications are sometimes relied upon as a means of data transmission. However, cosmic weather phenomena such as solar storms may interfere with satellite signals, resulting in large-scale communication interruptions. Satellite communication interruptions may last for hours or even days, affecting large areas of monitoring areas. Especially in extreme environmental monitoring (such as polar climate, marine pollution, etc.), long-term data transmission interruptions may result in key environmental events (such as climate change trends, marine pollution spread) not being tracked and reported in real time. At the same time, the lack of real-time data will cause the charts and maps in the visualization to be unable to update, and even present blank areas or outdated data. For example, in air quality monitoring, the distribution map of pollutants on the map does not reflect the latest situation, which may cause the public or decision makers to misunderstand the current environmental conditions. If a surge in pollution during a certain period of time is not captured in time, it may delay emergency response and endanger public health. Summary of the invention

[0006] The purpose of the present invention is to provide a platform and method for visualizing and displaying environmental monitoring data, so as to solve the deficiencies in the background technology.

[0007] In order to achieve the above object, the present invention provides the following technical solution: a method for visualizing environmental monitoring data, comprising the following steps:

[0008] S1: Divide the monitoring area into several sub-monitoring areas, and deploy different types of environmental monitoring equipment in each sub-monitoring area to collect environmental data in N time periods in real time, and transmit the environmental data collected in each time period to the central server through the communication network;

[0009] S2: The central server is used to receive and preprocess the environmental data collected in each time period, and perform feature extraction on the preprocessed environmental data, respectively extracting the transmission delay feature and the environmental data loss feature of the environmental data;

[0010] S3: The transmission delay characteristics and environmental data loss characteristics extracted in each time period are converted into a comprehensive feature vector as the input of the machine learning model, and the machine learning model is trained. The weight coefficients in different time periods during the environmental data transmission process are determined according to the model output results, and the weight coefficients of the environmental data in different time periods are weighted averaged to calculate the real-time index of environmental data transmission;

[0011] S4: Compare the calculated real-time index of environmental data transmission with the gradient standard threshold, divide the environmental data into real-time data, medium real-time data and non-real-time data, and directly use the real-time data to generate visualization charts and maps for environmental monitoring. In the visualization interface, non-real-time data should be clearly marked as abnormal data;

[0012] S5: For medium real-time data, conduct in-depth analysis on the real-time nature of environmental data transmission within a fixed time period, adjust the use scope of the medium real-time data based on the analysis results, and display the medium real-time data with different use scopes by region.

[0013] Preferably, in S2, the transmission delay characteristics of the extracted environmental data are analyzed to generate a transmission delay interference index, and the method for obtaining the transmission delay interference index is:

[0014] Collect transmission delay feature data in different time periods, including delay time, signal strength, packet loss rate, and bandwidth, and convert the historical transmission delay feature data set into Constructed as a basis vector matrix, where D is the basis vector matrix, containing n historical data samples, each sample contains m features, represents the column vector of the basis vector matrix, is the feature vector of historical features; the transmission delay feature at the current moment is It is represented by a sparse linear combination of the basis vector matrix D, ; In the formula, α∈ is a sparse coefficient vector, is the error term; the sparse coefficient vector α makes the delay feature y represented by the linear combination of the basis vector matrix D, expressed as: ; In the formula, is the reconstruction error, It is α norm, λ is the regularization parameter. After solving α, the reconstruction error RE of the current delayed feature vector y is calculated, and the expression is: ; Calculate the sparsity Sq, the expression is: ; Combine the reconstruction error and sparsity to calculate the transmission delay interference index, the expression is: ; In the formula, is the transmission delay interference index.

[0015] Preferably, in S2, the extracted environmental data loss characteristics are analyzed to generate a data loss anomaly index, and the method for obtaining the data loss anomaly index is:

[0016] Extract relevant data loss features from the environmental data transmission, where the data loss features include signal strength data and network bandwidth data, and construct a feature vector for each time period. Select k nearest neighbors, k determines the number of neighbors used to calculate the local density, and the selection of neighbors is calculated using the Euclidean distance metric. ;in There are two time periods and The distance between the feature vectors, n is the feature dimension, for each point , calculate the reachable distance between it and its nearest neighbor point k , the expression is: ; In the formula, Yes The distance to its kth nearest neighbor, Yes and The Euclidean distance between them; for each data point , its local reachability density is defined as the reciprocal mean of the reachability distances of all its neighboring points, expressed as: ; In the formula, Yes The k nearest neighbor set of is the number of neighbor points;

[0017] By comparing the same time period The local reachable density and the local reachability density of its neighbors , get the local outlier , the expression is: ; Normalize the local outliers of all time periods to generate the data loss anomaly index, which is expressed as: ; Where max(LOF) is the maximum local outlier in all time periods, and KH is the data loss anomaly index.

[0018] Preferably, in S3, the transmission delay characteristics of the environmental data extracted in each time period and the environmental data loss characteristics are converted into a comprehensive feature vector as an input of the machine learning model, and the machine learning model is trained, specifically:

[0019] The transmission delay interference index and the data loss anomaly index are converted into comprehensive feature vectors, and the comprehensive feature vectors are used as the input of the machine learning model. The machine learning model uses each group of comprehensive feature vectors to predict the weight coefficient value labels in different time periods during the environmental data transmission process as the prediction target, and takes minimizing the sum of prediction errors of the weight coefficient value labels in different time periods during all environmental data transmission processes as the training target. The machine learning model is trained until the sum of prediction errors converges and the model training is stopped. The weight coefficient values ​​in different time periods during the environmental data transmission process are determined according to the model output results. Among them, the machine learning model is a polynomial regression model, and the real-time index of environmental data transmission is calculated by weighted average calculation of the weight coefficients of environmental data in different time periods.

[0020] Preferably, in S4, the calculated real-time index of the environmental data transmission is compared with the gradient standard threshold, and the environmental data is divided into real-time data, medium real-time data and non-real-time data;

[0021] The real-time index of the environmental data transmission obtained is compared with the gradient standard threshold, the gradient standard threshold includes a first standard threshold and a second standard threshold, and the first standard threshold is less than the second standard threshold, and the real-time index of the environmental data transmission is compared with the first standard threshold and the second standard threshold respectively;

[0022] If the real-time index of environmental data transmission is greater than the second standard threshold, it means that the real-time performance of environmental data transmission is high. At this time, a high real-time signal is generated, and the environmental data is classified as real-time data; if the real-time index of environmental data transmission is greater than or equal to the first standard threshold and less than or equal to the second standard threshold, it means that the real-time performance of environmental data transmission is average. At this time, a medium real-time signal is generated, and the environmental data is classified as medium real-time data; if the real-time index of environmental data transmission is less than the first standard threshold, it means that the real-time performance of environmental data transmission is low. At this time, a low real-time signal is generated, and the environmental data is classified as low real-time data.

[0023] Preferably, in S5, for the medium real-time data, the real-time performance of the environmental data transmission within a fixed time period is deeply analyzed, and the use scope of the medium real-time data is adjusted according to the analysis result;

[0024] For medium real-time data, that is, the real-time index of environmental data transmission generated within a fixed time period is greater than or equal to the first standard threshold and less than or equal to the second standard threshold, the real-time index of environmental data transmission generated within a subsequent fixed time period that is greater than or equal to the first standard threshold and less than or equal to the second standard threshold is collected, and a corresponding data set is established, the mean and standard deviation of the data set are calculated, and after in-depth analysis, the use scope of the medium real-time data is adjusted according to the analysis results, and the medium real-time data with different use scopes are displayed in different regions.

[0025] Preferably, if the mean of the real-time index in the data set is greater than or equal to the reference threshold of the mean of the real-time index, and the standard deviation of the real-time index is less than the reference threshold of the standard deviation of the real-time index, it means that the mean of the data is high and the real-time preference is high; at the same time, the standard deviation is low, indicating that the data fluctuation is small and the data is stable; the data in the area is marked as high priority;

[0026] If the mean of the real-time index is greater than or equal to the reference threshold of the mean of the real-time index, and the standard deviation of the real-time index is greater than or equal to the reference threshold of the standard deviation of the real-time index, it means that the mean of the data is high and the real-time performance is good; but the standard deviation is large, which means that the data fluctuates greatly and is unstable, and the data in the area is marked as medium priority;

[0027] If the mean of the real-time index is less than the reference threshold of the real-time index mean, and the standard deviation of the real-time index is greater than or equal to the reference threshold of the real-time index standard deviation, the mean of the data is low and the real-time performance is poor; at the same time, the standard deviation is large, indicating that the data is highly volatile and the data quality is poor, and the data in the area is marked as low priority;

[0028] If the mean of the real-time index is less than the reference threshold of the mean of the real-time index, and the standard deviation of the real-time index is less than the reference threshold of the standard deviation of the real-time index, it means that the mean of the data is low and the real-time performance is low; but the standard deviation is small, which means that the data is stable, and the data in the area is marked as low priority.

[0029] The present invention also provides an environmental monitoring data visualization display platform, including a monitoring area division module, a feature extraction module, a real-time analysis module, a data classification module and a display module;

[0030] Monitoring area division module: divides the monitoring area into several sub-monitoring areas, and deploys different types of environmental monitoring equipment in each sub-monitoring area to collect environmental data in N time periods in real time, and transmits the environmental data collected in each time period to the central server through the communication network;

[0031] Feature extraction module: The central server is used to receive and preprocess the environmental data collected in each time period, and perform feature extraction on the preprocessed environmental data, respectively extracting the transmission delay feature of the environmental data and the environmental data loss feature;

[0032] Real-time analysis module: The transmission delay characteristics and environmental data loss characteristics extracted in each time period are converted into comprehensive feature vectors as the input of the machine learning model, and the machine learning model is trained. The weight coefficients in different time periods during the environmental data transmission process are determined according to the model output results, and the weight coefficients of the environmental data in different time periods are weighted averaged to calculate the real-time index of environmental data transmission;

[0033] Data classification module: Compare the calculated real-time index of environmental data transmission with the gradient standard threshold, divide the environmental data into real-time data, medium real-time data and non-real-time data, and directly use the real-time data to generate visualization charts and maps for environmental monitoring. In the visualization interface, non-real-time data should be clearly marked as abnormal data;

[0034] Display module: For medium real-time data, conduct in-depth analysis on the real-time nature of environmental data transmission within a fixed time period, adjust the use scope of the medium real-time data based on the analysis results, and display the medium real-time data with different use scopes in different regions.

[0035] In the above technical solution, the technical effects and advantages provided by the present invention are:

[0036] 1. By dividing the monitoring area into several sub-areas and deploying different types of monitoring equipment, the system can collect and transmit environmental data in real time to the central server for preprocessing, extract transmission delay and data loss characteristics, and use machine learning models to analyze these characteristics to calculate the real-time index, thereby classifying and displaying environmental data. High-real-time data is directly used for visualization, non-real-time data is marked as abnormal data, and medium-real-time data is further analyzed, and the use range is adjusted according to its stability and displayed in different regions.

[0037] 2. The present invention can not only dynamically evaluate and adjust the scope of use of environmental data, but also effectively ensure that the environmental monitoring system can still provide accurate and timely decision support in the event of communication interruption or data delay, avoiding the risk of key environmental events not being captured or reported in a timely manner. By using machine learning technology, the priority of data display can be automatically adjusted to ensure that high-priority data is used for real-time monitoring, while low-priority data is used for historical analysis and trend tracking, thereby improving the overall performance and reliability of the environmental monitoring system, especially in extreme environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0039] Figure 1 The present invention is a flow chart of the method.

[0040] Figure 2 It is a system module diagram of the present invention. DETAILED DESCRIPTION

[0041] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are 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.

[0042] Example 1, please refer to Figure 1 and Figure 2 As shown, the method for visualizing environmental monitoring data described in this embodiment includes the following steps:

[0043] S1: Divide the monitoring area into several sub-monitoring areas, and deploy different types of environmental monitoring equipment in each sub-monitoring area to collect environmental data in N time periods in real time, and transmit the environmental data collected in each time period to the central server through the communication network;

[0044] S2: The central server is used to receive and preprocess the environmental data collected in each time period, and perform feature extraction on the preprocessed environmental data, respectively extracting the transmission delay feature and the environmental data loss feature of the environmental data;

[0045] S3: The transmission delay characteristics and environmental data loss characteristics extracted in each time period are converted into a comprehensive feature vector as the input of the machine learning model, and the machine learning model is trained. The weight coefficients in different time periods during the environmental data transmission process are determined according to the model output results, and the weight coefficients of the environmental data in different time periods are weighted averaged to calculate the real-time index of environmental data transmission;

[0046] S4: Compare the calculated real-time index of environmental data transmission with the gradient standard threshold, divide the environmental data into real-time data, medium real-time data and non-real-time data, and directly use the real-time data to generate visualization charts and maps for environmental monitoring. In the visualization interface, non-real-time data should be clearly marked as abnormal data;

[0047] S5: For medium real-time data, conduct in-depth analysis on the real-time nature of environmental data transmission within a fixed time period, adjust the use scope of the medium real-time data based on the analysis results, and display the medium real-time data with different use scopes by region.

[0048] In S1, the monitoring area is divided into several sub-monitoring areas, and different types of environmental monitoring equipment are deployed in each sub-monitoring area to collect environmental data in N time periods in real time, and transmit the environmental data collected in each time period to the central server through the communication network, specifically:

[0049] Determine the geographical scope of remote areas that need to be monitored, such as mountainous areas, deserts, forests, oceans and other remote areas or areas that are difficult to access traditional communication networks. Divide the entire monitoring area into several sub-monitoring areas based on geographical characteristics, environmental factors and monitoring objectives. The basis for the division of each sub-area may include natural geographical boundaries (such as valleys, rivers), monitoring needs (such as the distribution of pollution sources), or the characteristics of satellite communication signal coverage. Each sub-area should be able to independently collect data and transmit it via satellite communication.

[0050] Determine the type of environmental data that needs to be monitored in each sub-monitoring area. For example, in the forest, it is necessary to monitor air quality, temperature and humidity, wind speed, etc.; near water bodies, it is necessary to monitor water quality (such as pH, dissolved oxygen, etc.) and water temperature. According to the specific environmental requirements of each sub-monitoring area, select appropriate environmental monitoring equipment. Optional equipment types include: air quality monitoring equipment: PM2.5, CO2, NO2 sensors, etc., meteorological monitoring equipment: temperature and humidity sensors, wind speed and direction instruments, air pressure sensors, water quality monitoring equipment: pH sensors, dissolved oxygen sensors, turbidity sensors, noise monitoring equipment: suitable for areas where noise pollution needs to be monitored. The equipment should have a satellite communication module to support the transmission of data to a remote central server via a satellite network. The communication module used is usually an L-band or C-band satellite communication module that supports low power consumption and long-distance transmission.

[0051] Deploy monitoring equipment at representative locations in each sub-monitoring area (such as mountains, rivers, vegetation areas, etc.). Ensure that each device can collect environmental data in the target area and is installed at a suitable height and location to avoid interference (such as obstruction and strong wind). Perform preliminary calibration on each monitoring device to ensure that the monitoring accuracy meets the requirements. For example, use standard gas to calibrate air quality sensors and use standard liquid to calibrate water quality sensors.

[0052] Set the collection frequency of the monitoring device, that is, automatically collect environmental data in N time periods (for example, every hour, every 6 hours). Make sure that the device can attach a timestamp, device ID, and geographic location information to each piece of data. Remote areas usually do not have ground communication infrastructure, so choosing the right satellite communication method is critical. The following satellite communication solutions can be selected: Low Earth Orbit (LEO): such as Starlink, suitable for transmitting large amounts of real-time monitoring data with low latency. Medium Earth Orbit (MEO): such as GPS satellites, used for data transmission with medium latency. Geostationary Earth Orbit (GEO): such as VSAT systems, suitable for covering wide areas, but with high latency, suitable for non-real-time data transmission.

[0053] Install satellite communication terminal equipment in each sub-monitoring area so that the monitoring equipment can transmit data via satellite communication. Ensure that the terminal equipment can cover the monitoring area and establish a stable connection with the satellite. Each monitoring device automatically collects environmental data such as air quality, water quality, temperature and humidity at a set frequency (such as every hour or every 6 hours). Perform preliminary preprocessing on the data in the local device: remove invalid data points and eliminate collection errors. In the case of limited bandwidth, the data can be compressed to reduce the transmission load. Attach a timestamp, sub-monitoring area identifier and device ID to each piece of data to ensure that the source can be correctly identified during data transmission and analysis.

[0054] Each monitoring device packages the data collected in N time periods through the satellite communication module and transmits it to the central server through the selected satellite communication network. The data can be transmitted in real time (such as every hour) or uploaded in batches (such as once a day). Appropriate transmission protocols are used in data transmission, such as: MQTT (Message Queuing TelemetryTransport): suitable for low-bandwidth and low-latency transmission. HTTP / HTTPS: suitable for high-bandwidth transmission requirements. FTP / SFTP: suitable for large files or batch data transmission.

[0055] S2: The central server is used to receive and preprocess the environmental data collected in each time period, and perform feature extraction on the preprocessed environmental data, respectively extracting the transmission delay feature of the environmental data and the environmental data loss feature.

[0056] The central server receives environmental data from each sub-monitoring area and stores it in a cloud or local database. The data is usually divided into real-time data and historical data, and stored by time and region. The received data is further cleaned to delete invalid or duplicate data packets to ensure data integrity. The data is formatted into a standard structure for subsequent analysis and visualization

[0057] The preprocessed environmental data is subjected to feature extraction, and the transmission delay feature and the environmental data loss feature of the environmental data are extracted respectively. The transmission delay feature of the extracted environmental data is analyzed to generate a transmission delay interference index. The method for obtaining the transmission delay interference index is as follows:

[0058] The transmission delay characteristic data in different time periods are collected. The characteristic data include: delay time, that is, the delay of data packets per unit time, signal strength, that is, the signal quality of the communication link, data packet loss rate, that is, the percentage of data packets lost per unit time, and bandwidth, that is, the amount of data transmitted per unit time.

[0059] The historical transmission delay characteristic dataset It is constructed as a basis vector matrix, where D is the basis vector matrix, containing n historical data samples, and each sample contains m features (such as delay, signal strength, packet loss rate, etc.). represents the column vector of the basis vector matrix, is the feature vector of historical features; the transmission delay feature at the current moment is It is represented by a sparse linear combination of the basis vector matrix D, ; In the formula, α∈ is a sparse coefficient vector, indicating that the delay feature can be represented by a linear combination of several basis vectors (sparseness means that many elements are close to 0), is the error term. The sparse coefficient vector α makes the delay feature y represented by the linear combination of the basis vector matrix D, expressed as: ; In the formula, is the reconstruction error (i.e., the squared error between the delayed signature and the reconstruction result), It is α norm is used to encourage sparsity, that is, the value of most sparse coefficients α is close to 0, and λ is a regularization parameter used to balance reconstruction error and sparsity. After solving α, the reconstruction error RE of the current delay feature vector y is calculated. The larger the reconstruction error, the more serious the transmission delay interference. The expression is: ; Calculate the sparsity Sq. Sparsity refers to the number of non-zero items in the sparse coefficient α. The more non-zero items there are, the more basis vectors can represent the delay feature, which means that the transmission delay is more disturbed. The expression is: ; Combine the reconstruction error and sparsity to calculate the transmission delay interference index, the expression is: ; In the formula, is the transmission delay interference index.

[0060] When the transmission delay interference index is large, it means that there is a large delay or anomaly in the transmission of environmental data. This usually means that the real-time performance of the transmission is poor, and the data transmission time is significantly longer than expected, which may be caused by network instability, limited bandwidth or high packet loss rate. Excessive transmission delay may cause the real-time monitoring data to be unable to be updated in time, and even make it impossible to capture key environmental events in time, thereby delaying decision-making and response actions.

[0061] On the contrary, when the transmission delay interference index is small, it indicates that the data transmission delay is small, the transmission process is stable, and the sparsity is high, indicating that the real-time performance of the transmission is good. This means that the monitoring equipment can transmit environmental data to the central server in a timely and reliable manner, ensuring the real-time performance of data updates. A transmission system with good real-time performance can ensure the continuous update of environmental data, so that the monitoring system can reflect the current environmental conditions in a timely manner, which helps to quickly respond to potential environmental changes or anomalies.

[0062] After analyzing the extracted environmental data loss characteristics, a data loss anomaly index is generated. The method for obtaining the data loss anomaly index is as follows:

[0063] Extract relevant data loss features from the environmental data transmission, where the data loss features include signal strength data and network bandwidth data, and construct a feature vector for each time period. Select k nearest neighbors, where k is the neighborhood radius parameter, and the common value range is 5 to 20. k determines the number of neighbors used to calculate the local density. The selection of neighbors is calculated using the Euclidean distance metric. ;in There are two time periods and The distance between the feature vectors, n is the feature dimension, for each point , calculate the reachable distance between it and its nearest neighbor point k , the expression is: ; In the formula, Yes The distance to its kth nearest neighbor, Yes and The Euclidean distance between them. The reachable distance represents With Neighbors distance, ensuring more robust processing of local outliers (i.e., processing areas with uneven density); for each data point , its local reachability density is defined as the reciprocal mean of the reachability distances of all its neighboring points, expressed as: ; In the formula, Yes The k nearest neighbor set of is the number of neighbor points.

[0064] By comparing the same time period The local reachable density and the local reachability density of its neighbors , get the local outlier The larger the local anomaly factor value is, the greater the difference between the characteristics of this time period and the adjacent time periods is, indicating that the data loss is abnormal. The expression is: ; Normalize the local outliers of all time periods to generate the data loss anomaly index, which is expressed as: ; Where max(LOF) is the maximum local outlier in all time periods, and KH is the data loss anomaly index.

[0065] When the data loss anomaly index is large, it means that a high level of data loss or damage occurred during the environmental data transmission process within a certain period of time. The larger the data loss anomaly index, the more serious the data loss and the worse the real-time transmission. This may mean that problems such as unstable network quality, weak signal strength, and frequent data packet loss make it impossible for environmental monitoring data to be transmitted to the central server on time or in full, which in turn affects the system's real-time monitoring capabilities, resulting in an inaccurate reflection of environmental conditions, which in turn delays decision-making and response.

[0066] On the contrary, when the data loss anomaly index is small, it means that there is very little loss or damage in the environmental data transmission process, and the transmission real-time performance is good. The data transmission process is stable, and data loss is rare or does not occur at all, ensuring that the data can be transmitted to the central server in a timely and complete manner. In this case, the system can maintain the ability of real-time monitoring, ensuring that decision makers and monitoring personnel can make correct judgments based on the latest data and effectively respond to environmental changes.

[0067] S3: The transmission delay characteristics and environmental data loss characteristics extracted in each time period are converted into a comprehensive feature vector as the input of the machine learning model, and the machine learning model is trained. The weight coefficients in different time periods during the environmental data transmission process are determined according to the model output results, and the weight coefficients of the environmental data in different time periods are weighted averaged to calculate the real-time index of environmental data transmission.

[0068] The transmission delay interference index and the data loss anomaly index are converted into comprehensive feature vectors, and the comprehensive feature vectors are used as the input of the machine learning model. The machine learning model uses each group of comprehensive feature vectors to predict the weight coefficient value labels in different time periods during the environmental data transmission process as the prediction target, and takes minimizing the sum of prediction errors of the weight coefficient value labels in different time periods during all environmental data transmission processes as the training target. The machine learning model is trained until the sum of prediction errors converges and the model training is stopped. The weight coefficient values ​​in different time periods during the environmental data transmission process are determined according to the model output results. Among them, the machine learning model is a polynomial regression model, and the real-time index of environmental data transmission is calculated by weighted average calculation of the weight coefficients of environmental data in different time periods.

[0069] The method for obtaining the weight coefficient values ​​in different time periods during the environmental data transmission process is: obtain the corresponding function expression from the first feature vector training data of the trained machine learning model: ; In the formula, is the output function of the model, is the transmission delay interference index, KH is the data loss anomaly index, It is the weight coefficient value in different time periods during the environmental data transmission process.

[0070] S4: Compare the calculated real-time index of environmental data transmission with the gradient standard threshold, divide the environmental data into real-time data, medium real-time data and non-real-time data, and directly use the real-time data to generate visualization charts and maps for environmental monitoring. In the visualization interface, non-real-time data should be clearly marked as abnormal data;

[0071] The real-time index of the environmental data transmission obtained is compared with the gradient standard threshold, the gradient standard threshold includes a first standard threshold and a second standard threshold, and the first standard threshold is less than the second standard threshold, and the real-time index of the environmental data transmission is compared with the first standard threshold and the second standard threshold respectively;

[0072] If the real-time index of environmental data transmission is greater than the second standard threshold, it means that the real-time performance of environmental data transmission is high. At this time, a high real-time signal is generated, and the environmental data is classified as real-time data; if the real-time index of environmental data transmission is greater than or equal to the first standard threshold and less than or equal to the second standard threshold, it means that the real-time performance of environmental data transmission is average. At this time, a medium real-time signal is generated, and the environmental data is classified as medium real-time data; if the real-time index of environmental data transmission is less than the first standard threshold, it means that the real-time performance of environmental data transmission is low. At this time, a low real-time signal is generated, and the environmental data is classified as low real-time data.

[0073] Real-time data is data with high real-time performance. This data can accurately and timely reflect the current status of the environment, so it can be directly used to generate charts and maps for environmental monitoring.

[0074] Parse the collected real-time data, including data formatting, timestamp processing, and extraction of geographic location information. Ensure that the data accurately reflects the time and location of collection and is ready for visualization. Use real-time data to generate line charts, bar charts, etc. to show environmental changes in different time periods. For example, show the changing trend of air quality (such as PM2.5 concentration changes over time) or the dynamic changes of water quality parameters (such as dissolved oxygen, pH value, etc.). For data with geospatial information, draw it on a map to generate a real-time environmental map. For example: Air quality map: Use color gradients (such as green for good and red for pollution) to display air quality data for various regions. Show the water quality of different locations on a map of water distribution, and use color or symbol size to show the degree of pollution.

[0075] Real-time data can be used to dynamically update the visualization interface, ensuring that charts and maps can continuously reflect the latest monitoring data. Whenever new real-time data arrives, the system automatically refreshes the visualization content to ensure that users are viewing the current environmental status.

[0076] Non-real-time data, also known as low-real-time data, usually cannot reflect the current environmental conditions in a timely manner due to transmission delays, data loss or other transmission anomalies. Therefore, such data should be clearly marked as abnormal in the visualization interface to remind users that these data may no longer reflect the actual situation.

[0077] According to the calculation results of the real-time index, non-real-time data has been identified and classified in the system. The system automatically marks these data as "low real-time" or "abnormal data". The non-real-time data is further evaluated through indicators such as the transmission delay interference index and the data loss abnormality index to confirm that it cannot reflect the latest environmental conditions. On the visual map, special colors, symbols or graphics are used to mark non-real-time data. Common marking methods include: Gray area: Gray indicates that the data in the area is outdated or invalid. For abnormal data in key areas (such as pollution sources and hazardous areas), red marks are used and warning symbols are attached, such as triangular warning icons. In the chart or map, the user is informed of the specific problems of the data through floating prompts. For example, when the user hovers the mouse over the abnormal area, the prompt box will display "The data is delayed for more than 5 hours and may not reflect the current situation."

[0078] For non-real-time data, in addition to abnormal marking, data delay and error information can also be provided in the visual interface. Users can view detailed information to understand the delay time or possible error range of the data. For example, it can be marked as "This data was collected 3 hours ago, and the current situation may have changed."

[0079] To ensure that users can obtain comprehensive environmental monitoring information, the system can display both real-time data and non-real-time data, but they must be clearly distinguished. Real-time data and non-real-time data are distinguished by different colors or transparency. For example, real-time data uses bright colors (such as green or blue), while non-real-time data uses gray or transparent colors to indicate that it is not timely. Next to each data point or area, the timestamp of the data is displayed. For example, in the environmental map, "Last updated: 10 minutes ago" is displayed next to the data of each monitoring point, while abnormal data may show "Last updated: 5 hours ago". In time series charts, non-real-time data can be marked with broken lines or dotted lines. For example, in a line chart showing changes in air quality, the part of the non-real-time data that exceeds the real-time threshold is marked with a dotted line or color change, so that users can clearly distinguish which data is real-time and which data may be outdated.

[0080] S5: For medium real-time data, conduct in-depth analysis on the real-time nature of environmental data transmission within a fixed time period, adjust the use scope of the medium real-time data based on the analysis results, and display the medium real-time data with different use scopes by region.

[0081] For medium real-time data, that is, the real-time index of environmental data transmission generated within a fixed time period is greater than or equal to the first standard threshold and less than or equal to the second standard threshold, the real-time index of environmental data transmission generated within a subsequent fixed time period that is greater than or equal to the first standard threshold and less than or equal to the second standard threshold is collected, and a corresponding data set is established, the mean and standard deviation of the data set are calculated, and after in-depth analysis, the use scope of the medium real-time data is adjusted according to the analysis results, and the medium real-time data with different use scopes are displayed in different regions.

[0082] If the mean of the real-time index in the data set is greater than or equal to the reference threshold of the mean of the real-time index, and the standard deviation of the real-time index is less than the reference threshold of the standard deviation of the real-time index, it means that the mean of the data is high and real-time preference is given; at the same time, the standard deviation is low, indicating that the data fluctuation is small and the data is stable; in this case, medium real-time data can be safely used in a wider range of application scenarios, such as dynamic monitoring and short-term prediction. In the visualization interface, the data in this area is marked as high priority and displayed in bright colors to prompt users that the data can be used for real-time monitoring.

[0083] If the mean of the real-time index is greater than or equal to the reference threshold of the mean of the real-time index, and the standard deviation of the real-time index is greater than or equal to the reference threshold of the standard deviation of the real-time index, it means that the mean of the data is high and the real-time performance is good; but the standard deviation is large, which means that the data fluctuates greatly and is unstable. The data in this area is marked as medium priority. In this case, the data can be used for monitoring and analysis, but the volatility of the data should be noted, and further smoothing or data compensation may be required. In the visualization interface, the data is marked with a medium color, and an additional prompt indicates that the data fluctuates. It is recommended to use it in combination with other data sources.

[0084] If the mean of the real-time index is less than the reference threshold of the mean of the real-time index, and the standard deviation of the real-time index is greater than or equal to the reference threshold of the standard deviation of the real-time index, the mean of the data is low and the real-time performance is poor; at the same time, the standard deviation is large, indicating that the data is highly volatile and the data quality is poor. Mark the data in this area as low priority. This data should be limited to non-real-time applications and only used for trend analysis or historical data review, and avoid use in real-time monitoring. Use light or transparent marks in the visual interface to remind users that the data has poor real-time performance and is not suitable for real-time applications.

[0085] If the mean of the real-time index is less than the reference threshold of the mean of the real-time index, and the standard deviation of the real-time index is less than the reference threshold of the standard deviation of the real-time index, it means that the mean of the data is low and the real-time performance is low; but the standard deviation is small, indicating that the data is relatively stable and has no significant fluctuations. Mark the data in this area as low priority. This data is suitable for non-critical monitoring tasks, such as long-term trend analysis or minor decision support. Use light-colored marks in the visualization interface to indicate that the data is stable but lacks real-time performance. It is recommended to use it only for analysis.

[0086] After adjusting the scope of use, data will be displayed in different regions according to different priorities and usage scenarios based on the real-time nature of the data in each region.

[0087] The data is divided into geographical regions, and the real-time data of each region can be analyzed independently for its mean and standard deviation, ensuring that the data display of different regions can reflect their respective real-time characteristics.

[0088] High priority area: For areas that meet the real-time index mean greater than or equal to the reference threshold of the real-time index mean, and the real-time index standard deviation greater than or equal to the reference threshold of the real-time index standard deviation, bright colors and dynamic charts are used to remind users that the data in this area can be used for real-time monitoring.

[0089] Medium priority area: For areas that meet the real-time index mean greater than or equal to the reference threshold of the real-time index mean, and the real-time index standard deviation greater than or equal to the reference threshold of the real-time index standard deviation, medium colors are used, and additional prompts are added to explain the possibility of data fluctuations.

[0090] Low priority area: For areas that meet the conditions that the real-time index mean is less than the reference threshold of the real-time index mean, and the real-time index standard deviation is greater than or equal to the reference threshold of the real-time index standard deviation, and the real-time index mean is less than the reference threshold of the real-time index mean, and the real-time index standard deviation is less than the reference threshold of the real-time index standard deviation, light colors or transparent effects are used to display the areas, prompting users that the data in these areas are only suitable for historical analysis or decision support.

[0091] In this embodiment, in the environmental monitoring system, the monitoring area is first divided into several sub-monitoring areas, and different types of monitoring equipment are deployed in each area to collect environmental data in multiple time periods in real time, and these data are transmitted to the central server through the communication network. After receiving the data, the central server performs preprocessing and feature extraction, extracts the transmission delay characteristics and data loss characteristics of the environmental data, and converts them into comprehensive feature vectors, which are trained as inputs of the machine learning model. According to the output results of the model, the weight coefficient of each time period is determined, and then the real-time index of environmental data transmission is calculated by weighted average. Then, the real-time index is compared with the gradient standard threshold, and the data is divided into real-time data, medium real-time data and non-real-time data. The real-time data is directly used to generate charts and maps for environmental monitoring, and the non-real-time data is marked as abnormal data in the visual interface. For medium real-time data, the real-time performance is further analyzed, its scope of use is adjusted, and the data is displayed in different regions based on different scopes of use.

[0092] Embodiment 2, the environmental monitoring data visualization display platform described in this embodiment includes a monitoring area division module, a feature extraction module, a real-time analysis module, a data classification module and a display module;

[0093] Monitoring area division module: divides the monitoring area into several sub-monitoring areas, and deploys different types of environmental monitoring equipment in each sub-monitoring area to collect environmental data in N time periods in real time, and transmits the environmental data collected in each time period to the central server through the communication network;

[0094] Feature extraction module: The central server is used to receive and preprocess the environmental data collected in each time period, and perform feature extraction on the preprocessed environmental data, respectively extracting the transmission delay feature of the environmental data and the environmental data loss feature;

[0095] Real-time analysis module: The transmission delay characteristics and environmental data loss characteristics extracted in each time period are converted into comprehensive feature vectors as the input of the machine learning model, and the machine learning model is trained. The weight coefficients in different time periods during the environmental data transmission process are determined according to the model output results, and the weight coefficients of the environmental data in different time periods are weighted averaged to calculate the real-time index of environmental data transmission;

[0096] Data classification module: Compare the calculated real-time index of environmental data transmission with the gradient standard threshold, divide the environmental data into real-time data, medium real-time data and non-real-time data, and directly use the real-time data to generate visualization charts and maps for environmental monitoring. In the visualization interface, non-real-time data should be clearly marked as abnormal data;

[0097] Display module: For medium real-time data, conduct in-depth analysis on the real-time nature of environmental data transmission within a fixed time period, adjust the use scope of the medium real-time data based on the analysis results, and display the medium real-time data with different use scopes in different regions.

[0098] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0099] It should be understood that the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. A and B can be singular or plural. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship, but it may also indicate an "and / or" relationship. Please refer to the context for specific understanding.

[0100] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0101] The above description is only a specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application.

Claims

1. A method for visualizing environmental monitoring data, characterized in that: The steps include: S1: Divide the monitoring area into several sub-monitoring areas, and deploy different types of environmental monitoring equipment in each sub-monitoring area to collect environmental data in N time periods in real time, and transmit the environmental data collected in each time period to the central server through the communication network; S2: The central server is used to receive and preprocess the environmental data collected in each time period, and perform feature extraction on the preprocessed environmental data, respectively extracting the transmission delay feature and the environmental data loss feature of the environmental data; S3: The transmission delay characteristics and environmental data loss characteristics extracted in each time period are converted into a comprehensive feature vector as the input of the machine learning model, and the machine learning model is trained. The weight coefficients in different time periods during the environmental data transmission process are determined according to the model output results, and the weight coefficients of the environmental data in different time periods are weighted averaged to calculate the real-time index of environmental data transmission; S4: Compare the calculated real-time index of environmental data transmission with the gradient standard threshold, divide the environmental data into real-time data, medium real-time data and non-real-time data, and directly use the real-time data to generate visualization charts and maps for environmental monitoring. In the visualization interface, non-real-time data should be clearly marked as abnormal data; S5: For medium real-time data, conduct an in-depth analysis of the real-time nature of environmental data transmission within a fixed time period, adjust the use scope of the medium real-time data based on the analysis results, and display the medium real-time data with different use scopes by region; The step S5 includes: for medium real-time data, that is, the real-time index of environmental data transmission generated within a fixed time period is greater than or equal to the first standard threshold and less than or equal to the second standard threshold, the real-time index of environmental data transmission generated within a subsequent fixed time period that is greater than or equal to the first standard threshold and less than or equal to the second standard threshold is collected, and a corresponding data set is established, the mean and standard deviation of the data set are calculated, and after in-depth analysis, the use scope of the medium real-time data is adjusted according to the analysis results, and the medium real-time data with different use scopes are displayed by region; If the mean of the real-time index in the data set is greater than or equal to the reference threshold of the real-time index mean, and the standard deviation of the real-time index is less than the reference threshold of the real-time index standard deviation, it means that the mean of the data is high and real-time preference is given; at the same time, the standard deviation is low, indicating that the data fluctuation is small and the data is stable; the data in the area is marked as high priority; If the mean of the real-time index is greater than or equal to the reference threshold of the mean of the real-time index, and the standard deviation of the real-time index is greater than or equal to the reference threshold of the standard deviation of the real-time index, it means that the mean of the data is high and the real-time performance is good; but the standard deviation is large, which means that the data fluctuates greatly and is unstable, and the data in the area is marked as medium priority; If the mean of the real-time index is less than the reference threshold of the real-time index mean, and the standard deviation of the real-time index is greater than or equal to the reference threshold of the real-time index standard deviation, the mean of the data is low and the real-time performance is poor; at the same time, the standard deviation is large, indicating that the data is highly volatile and the data quality is poor, and the data in the area is marked as low priority; If the mean of the real-time index is less than the reference threshold of the mean of the real-time index, and the standard deviation of the real-time index is less than the reference threshold of the standard deviation of the real-time index, it means that the mean of the data is low and the real-time performance is low; but the standard deviation is small, which means that the data is stable, and the data in the area is marked as low priority.

2. The method for visualizing environmental monitoring data according to claim 1, characterized in that: In S2, the transmission delay characteristics of the extracted environmental data are analyzed to generate a transmission delay interference index. The method for obtaining the transmission delay interference index is: Collect transmission delay feature data in different time periods, including delay time, signal strength, packet loss rate and bandwidth, and convert historical transmission delay feature data sets into Constructed as a basis vector matrix, where D is the basis vector matrix, containing Q historical data samples, each sample contains P features, represents the column vector of the basis vector matrix, is the feature vector of historical features; the transmission delay feature at the current moment is It is represented by a sparse linear combination of the basis vector matrix D, ; In the formula, α∈ is a sparse coefficient vector, is the error term; the sparse coefficient vector α makes the delay feature y represented by the linear combination of the basis vector matrix D, expressed as: ; In the formula, is the reconstruction error, It is α norm, λ is the regularization parameter. After solving α, the reconstruction error RE of the current delayed feature vector y is calculated, and the expression is: ; Calculate the sparsity Sq, the expression is: ; Combine the reconstruction error and sparsity to calculate the transmission delay interference index, the expression is: ; In the formula, is the transmission delay interference index.

3. The method for visualizing environmental monitoring data according to claim 2, characterized in that: In S2, the extracted environmental data loss characteristics are analyzed to generate a data loss anomaly index. The method for obtaining the data loss anomaly index is: Extract relevant data loss features from the environmental data transmission, where the data loss features include signal strength data and network bandwidth data, and generate data points for each time period. Select k nearest neighbors, k determines the number of neighbors used to calculate the local density, and the selection of neighbors is calculated using the Euclidean distance metric. ;in It is a data point and The distance between the feature vectors of , calculate the reachable distance between it and its nearest neighbor point k , the expression is: ; In the formula, It is a data point The distance to its kth nearest neighbor, It is a data point and The Euclidean distance between , its local reachability density is defined as the reciprocal mean of the reachability distances of all its neighboring points, expressed as: ; In the formula, It is a data point The k nearest neighbor set of is the number of neighbor points; By comparing data points within the same time period The local reachable density and the local reachability density of its neighbors , get the local outlier , the expression is: ; Normalize the local outliers of all time periods to generate the data loss anomaly index, which is expressed as: ; Where max(LOF) is the maximum local outlier in all time periods, and KH is the data loss anomaly index.

4. The method for visualizing environmental monitoring data according to claim 3, characterized in that: In S3, the transmission delay characteristics of the environmental data extracted in each time period and the environmental data loss characteristics are converted into a comprehensive feature vector as the input of the machine learning model, and the machine learning model is trained, specifically: The transmission delay interference index and the data loss anomaly index are converted into comprehensive feature vectors, and the comprehensive feature vectors are used as the input of the machine learning model. The machine learning model uses each group of comprehensive feature vectors to predict the weight coefficient value labels in different time periods during the environmental data transmission process as the prediction target, and takes minimizing the sum of prediction errors of the weight coefficient value labels in different time periods during all environmental data transmission processes as the training target. The machine learning model is trained until the sum of prediction errors converges and the model training is stopped. The weight coefficient values ​​in different time periods during the environmental data transmission process are determined according to the model output results. Among them, the machine learning model is a polynomial regression model, and the real-time index of environmental data transmission is calculated by weighted average calculation of the weight coefficients of environmental data in different time periods.

5. The method for visualizing environmental monitoring data according to claim 4, characterized in that: In S4, the calculated real-time index of environmental data transmission is compared with the gradient standard threshold, and the environmental data is divided into real-time data, medium real-time data and non-real-time data; The real-time index of the environmental data transmission obtained is compared with the gradient standard threshold, the gradient standard threshold includes a first standard threshold and a second standard threshold, and the first standard threshold is less than the second standard threshold, and the real-time index of the environmental data transmission is compared with the first standard threshold and the second standard threshold respectively; If the real-time index of environmental data transmission is greater than the second standard threshold, it means that the real-time performance of environmental data transmission is high. At this time, a high real-time signal is generated, and the environmental data is classified as real-time data; if the real-time index of environmental data transmission is greater than or equal to the first standard threshold and less than or equal to the second standard threshold, it means that the real-time performance of environmental data transmission is average. At this time, a medium real-time signal is generated, and the environmental data is classified as medium real-time data; if the real-time index of environmental data transmission is less than the first standard threshold, it means that the real-time performance of environmental data transmission is low. At this time, a low real-time signal is generated, and the environmental data is classified as low real-time data.

6. An environmental monitoring data visualization display platform, used to implement the environmental monitoring data visualization display method according to any one of claims 1 to 5, characterized in that: It includes monitoring area division module, feature extraction module, real-time analysis module, data classification module and display module; Monitoring area division module: divides the monitoring area into several sub-monitoring areas, and deploys different types of environmental monitoring equipment in each sub-monitoring area to collect environmental data in N time periods in real time, and transmits the environmental data collected in each time period to the central server through the communication network; Feature extraction module: The central server is used to receive and preprocess the environmental data collected in each time period, and perform feature extraction on the preprocessed environmental data, respectively extracting the transmission delay feature of the environmental data and the environmental data loss feature; Real-time analysis module: The transmission delay characteristics and environmental data loss characteristics extracted in each time period are converted into comprehensive feature vectors as the input of the machine learning model, and the machine learning model is trained. The weight coefficients in different time periods during the environmental data transmission process are determined according to the model output results, and the weight coefficients of the environmental data in different time periods are weighted averaged to calculate the real-time index of environmental data transmission; Data classification module: Compare the calculated real-time index of environmental data transmission with the gradient standard threshold, divide the environmental data into real-time data, medium real-time data and non-real-time data, and directly use the real-time data to generate visualization charts and maps for environmental monitoring. In the visualization interface, non-real-time data should be clearly marked as abnormal data; Display module: For medium real-time data, conduct in-depth analysis on the real-time nature of environmental data transmission within a fixed time period, adjust the use scope of the medium real-time data based on the analysis results, and display the medium real-time data with different use scopes in different regions.

Images