Regional environment high-precision monitoring system based on communication tower group

Abstract

The invention discloses a regional environment high-precision monitoring system based on a communication tower group. The regional environment high-precision monitoring system comprises an environment monitoring module, a data transmission module, a data center, a data analysis and early warning platform and a user side, wherein the environment monitoring module is used for monitoring environment parameter data in real time through a plurality of sensors; the data transmission module is used for transmitting the acquired environment parameter data among a plurality of communication towers in the area and transmitting the environment parameter data to a data center; the data center is used for collecting and storing the environment monitoring data transmitted by the data transmission module; the data analysis and early warning platform is used for acquiring corresponding data from the data center according to actual requirements for data analysis, feeding back an analysis result and early warning information to the data center for storage, and sending the early warning information to related management personnel; and the user side is used for providing a Web page and a friendly visual interface of the mobile side for a user.

Description

Technical Field

[0001] The present invention relates to the technical field of environmental monitoring, and in particular to a high-precision regional environment monitoring system based on a communication tower group. Background Art

[0002] With the enhancement of environmental protection awareness, the demand for monitoring environmental factors such as atmosphere, water quality, and noise is growing. Traditional environmental monitoring stations are usually set up in representative locations, such as city centers, industrial areas, residential areas, nature reserves, etc., to ensure that the environmental quality conditions in different regions can be fully reflected, and different monitoring equipment or instruments need to be set up to measure various environmental factors such as air, water quality, and soil. It can be seen that these environmental monitoring stations are sparsely distributed, the monitored data is single and limited, and the construction and maintenance costs of traditional environmental monitoring stations are high. Therefore, how to design a system that can monitor environmental parameters in all directions is a problem that needs to be solved urgently. The present invention utilizes the widely distributed communication towers in the city, and through its height and coverage advantages, integrates environmental monitoring functions to achieve continuous, real-time, and high-precision monitoring of environmental parameters in a wide range of areas. Summary of the invention

[0003] The purpose of the present invention is to provide a high-precision monitoring system for regional environment based on a communication tower group, which aims to utilize the existing communication tower infrastructure, high-precision sensors and intelligent analysis methods to achieve continuous, real-time and high-precision monitoring of environmental parameters.

[0004] To achieve the above objectives, the present invention provides a high-precision regional environment monitoring system based on a communication tower group, which is deployed on the communication towers in the area to be monitored, and includes:

[0005] Environmental monitoring module, used to monitor environmental parameter data in real time through multiple sensors;

[0006] A data transmission module, used to transmit the monitored environmental parameter data between multiple communication towers in the area and transmit the environmental parameter data to a data center;

[0007] A data center, used to collect and store environmental monitoring data transmitted by the data transmission module;

[0008] The data analysis and early warning platform is used to obtain the corresponding data from the data center for data analysis according to actual needs, and feed back the analysis results and early warning information to the data center for storage, and send the early warning information to relevant managers;

[0009] The user side, including the web side and mobile APP, is used to provide users with web pages and mobile-friendly visualization interfaces, visually display the distribution of monitoring points, real-time environmental data query, historical data analysis and comparison, and provide an environmental event reporting interface for the public to report environmental events through the APP.

[0010] Furthermore, the environment monitoring module includes:

[0011] An air quality monitoring unit, used for monitoring the concentration of main pollutants in the air in real time through a first sensor group;

[0012] A meteorological parameter monitoring unit, used for real-time monitoring of temperature, humidity, wind speed and direction, and air pressure through a second sensor group;

[0013] A noise pollution monitoring unit is used to collect current environmental noise data through acoustic and electrical sensors and obtain the level of noise pollution according to noise standards;

[0014] Special environmental factor monitoring unit, used to monitor radiation dose, water quality, air pollution, soil quality, noise level, biodiversity and some specific pollutants through the optional third sensor group.

[0015] Further, the first sensor group includes a PM2.5 / PM10 particulate matter sensor, a sulfur dioxide gas sensor, a nitrogen dioxide gas sensor, a carbon monoxide gas sensor and an ozone gas sensor;

[0016] The second sensor group includes a humidity sensor, an anemometer, and a barometer;

[0017] The third sensor group includes a radiation dosimeter and a water quality monitoring probe.

[0018] Furthermore, the data transmission module includes:

[0019] A wireless transmission unit, used to transmit the environmental parameter data monitored by the environmental monitoring module to the data center via a low-power wide area network;

[0020] The data encryption unit is used to perform AES encryption transmission on the original environment parameter data.

[0021] Furthermore, the data center includes:

[0022] Distributed storage database for storing and retrieving collected data;

[0023] The elastic computing module is used to utilize the elastic computing service provided by the cloud service provider to determine the normal and peak load data of the monitoring system, wherein the load data includes CPU usage, memory requirements, storage space and network bandwidth.

[0024] Furthermore, the data analysis and early warning platform includes:

[0025] Advanced data analysis module for identifying abnormal fluctuations in environmental parameters based on current environmental data and historical data;

[0026] The real-time warning and response module is used to issue an alarm based on the detection results of the anomaly detection unit and send response suggestions to the operating user.

[0027] Furthermore, the advanced data analysis module includes:

[0028] An environment change model building unit is used to build an environment change model, provide prediction data to the anomaly detection unit through the environment change model, and assist the anomaly detection unit in analyzing data trends;

[0029] The anomaly detection unit, based on historical and real-time monitored environmental data, first uses statistical methods to quickly screen out obvious outliers, then uses deep learning algorithms to identify abnormal fluctuations in environmental parameters, and feeds the abnormal data back to the environmental change model to update the model parameters.

[0030] Furthermore, the real-time warning and response module includes:

[0031] Threshold setting unit, used to set the alert and warning thresholds of various environmental indicators according to national environmental standards and local regulations;

[0032] An early warning unit is used to send warning information to the operating user;

[0033] The response suggestion unit is used to generate preliminary emergency response suggestions based on the type of abnormal event.

[0034] Furthermore, the step of constructing the environmental change model includes:

[0035] Step 1: Collect historical and current environmental data according to the type of environmental change to be monitored, and pre-process the collected environmental data;

[0036] Step 2: Select the appropriate model type in the advanced data analysis module according to actual needs, set model parameters and variable factors, and input the data preprocessed in step 1 into the selected model to train and optimize the model;

[0037] Step 3: Use the optimized model to simulate future environmental changes, generate prediction results based on different scenarios, and perform uncertainty analysis;

[0038] Step 4: Compare the predicted results with the actual situation or subsequent observation data, evaluate the model, and adjust the model parameters according to the evaluation results until the model meets the preset error threshold and outputs the constructed environmental change model.

[0039] Furthermore, the specific steps of the abnormal detection unit using a statistical method to screen abnormal values ​​of environmental parameters include:

[0040] Step 1: Obtain historical and real-time monitoring environmental data;

[0041] Step 2: Perform statistical analysis on the data through box plot analysis to obtain outliers;

[0042] Step 2.1.1: Draw a box plot. For each environmental parameter, draw a box plot separately;

[0043] Step 2.1.2: Determine the quartiles, calculate the first quartile Q1, the third quartile Q3, and the interquartile range IQR = Q3-Q1;

[0044] Step 2.1.3: According to the box plot rules, any data point less than Q1-1.5*IQR or greater than Q3+1.5*IQR is considered an outlier;

[0045] Step 2.1.4: Record and mark all outliers;

[0046] Step 3: Based on the results obtained in step 2, the environmental data are analyzed using the Z-score method to obtain standardized data;

[0047] Step 3.1.1: Calculate the mean and standard deviation: For each environmental parameter, calculate the mean μ and standard deviation σ of its historical data;

[0048] Step 3.1.2: Record the data points with Z-score absolute value greater than 3 as outliers;

[0049] Step 3.1.3: Convert all data points to Z-score to obtain a standardized data set;

[0050] Step 4: Clean the standardized data to obtain a clean environmental data set;

[0051] Step 4.1.1: Handling outliers;

[0052] Specifically, outliers are deleted or corrected using interpolation methods, and all deleted outliers and their causes are recorded;

[0053] Step 4.1.2: Use sliding window averaging to reduce noise;

[0054] Step 4.1.3: Verify the processed data and output the final historical and real-time environmental datasets.

[0055] Therefore, the present invention adopts the above-mentioned regional environment high-precision monitoring system based on communication tower groups, which has the following beneficial effects:

[0056] First, the present invention can achieve real-time monitoring of environmental parameters such as air quality, meteorological parameters, and noise pollution by integrating high-precision sensors on the existing communication tower infrastructure, and transmit the collected data using the communication tower;

[0057] Second, the present invention sets up a data analysis and early warning platform, which can use machine learning models to analyze and simulate environmental parameters, and issue real-time alarms after abnormal data is detected.

[0058] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 This is the module architecture diagram of this system. DETAILED DESCRIPTION

[0060] In the description of the present invention, it is also necessary to explain that, unless otherwise clearly specified and limited, these embodiments are only used to illustrate the present invention and are not used to limit the scope of the present invention. In addition, it should be understood that after reading the content taught by the present invention, those skilled in the art make various changes or modifications to the present invention, and these equivalent forms also fall within the scope limited by the claims attached to the application.

[0061] The present invention discloses a regional environment high-precision monitoring system based on a communication tower group, which is deployed on the existing communication tower infrastructure in the monitoring area to achieve all-round, high-density, real-time monitoring of environmental parameters in the area, improve monitoring efficiency and accuracy, and reduce monitoring costs.

[0062] It includes the following modules:

[0063] 1. Environmental monitoring module

[0064] This module is used to monitor various environmental parameters in real time through multiple sensors. It includes the following sub-modules:

[0065] (1) Air quality monitoring unit

[0066] By integrating highly sensitive PM2.5 / PM10 particulate matter sensors, sulfur dioxide (SO2), nitrogen dioxide (NO2), carbon monoxide (CO) and ozone (O3) gas sensors, it can be used to monitor the concentration of major pollutants in the air in real time;

[0067] (2) Meteorological parameter monitoring unit

[0068] By integrating humidity sensors, anemometers and barometers, it can capture temperature, humidity, wind speed and direction, and air pressure;

[0069] (3) Noise pollution monitoring unit

[0070] By setting up acoustic and electrical sensors to collect current environmental noise data, it is used to evaluate the level of noise pollution and to evaluate the noise pollution status according to daytime and nighttime standards. In China, the evaluation standards for noise pollution are usually based on the "Environmental Noise Pollution Prevention and Control Law" and relevant national standards, such as the "Urban Area Environmental Noise Standard" (GB 3096-2008).

[0071] Different regions and countries have different regulations for daytime and nighttime noise standards, but they are usually divided into several categories:

[0072] Category 1 area: refers to areas dominated by residences and cultural and educational institutions.

[0073] Category 2 area: refers to mixed residential, commercial and industrial areas.

[0074] Category 3 areas: refers to areas dominated by industrial production, warehousing and logistics.

[0075] Category 4 area: refers to the areas on both sides of the main road traffic lines.

[0076] The noise limits for each category are different. Generally speaking, the noise limits at night are stricter than those during the day because people need a quiet environment to rest at night. For example, in a Category 1 area, the noise limit during the day may be 55 decibels (A sound level, referred to as dB(A)), while the noise limit at night is reduced to 45 decibels.

[0077] The quantification of noise pollution is usually expressed in dB(A), a frequency-weighted sound pressure level measurement that better reflects the sensitivity of the human ear to sounds of different frequencies. Noise monitoring systems continuously record noise levels over a period of time and calculate noise pollution statistics based on these data.

[0078] (4) Special environmental factor monitoring unit

[0079] According to regional characteristics, radiation dosimeters and water quality monitoring probes can be optionally configured to monitor radiation dose, water quality, air pollution, soil quality, noise level, biodiversity and some specific pollutants to meet the monitoring needs of specific areas.

[0080] The present invention ensures long-term stable operation of multiple sensors through a complementary power supply solution of solar panels or wind energy on communication tower facilities, combined with a low power consumption design.

[0081] Since all of the above sensors have a self-calibration mechanism, they can regularly compare data with remote reference stations and automatically adjust measurement parameters to reduce errors caused by environmental changes and equipment aging.

[0082] 2. Data transmission module

[0083] This module is used to transmit the environmental parameters monitored by the environmental monitoring module between multiple communication towers in the area, and finally transmit the environmental parameter data to the data center; it specifically includes the following sub-modules:

[0084] (1) Wireless transmission unit

[0085] The wireless transmission unit is used to transmit the environmental parameters monitored by the environmental monitoring module to the data center through a low-power wide area network (LPWAN) such as LoRa, Sigfox or NB-IoT. The LPWAN network can achieve long-distance, low-power data transmission, and can ensure reliable data transmission even in remote areas or areas with poor signals;

[0086] (2) Data encryption unit

[0087] Data encryption unit, used to encrypt the monitored environmental parameters, using AES encryption during data transmission to ensure the security and privacy protection of environmental monitoring data;

[0088] 3. Data Center

[0089] The data center is used to collect and store data from various environmental monitoring (such as air quality monitoring, water quality monitoring, etc.), including:

[0090] (1) Distributed Storage Database

[0091] Used to store and retrieve collected data to ensure data integrity and availability;

[0092] (2) Elastic computing module

[0093] Use the elastic computing services provided by cloud service providers to determine the normal and peak load conditions of this monitoring system, including CPU usage, memory requirements, storage space, and network bandwidth;

[0094] Select the appropriate instance type according to the characteristics exhibited during system operation, such as general-purpose, compute-optimized, memory-optimized, storage-optimized, and CPU-optimized. Create at least one basic server instance as a template on the cloud service provider's console. Set up an elastic scaling group and configure an elastic scaling policy in the cloud service provider's elastic scaling service, enable the monitoring service provided by the cloud service provider, track key indicators, set alarm rules according to business needs, and receive notifications when resource usage reaches the preset threshold. Then conduct a simulated load test and continuously adjust the scaling policy based on the test results and actual operating data to ensure efficient resource utilization.

[0095] 4. Data analysis and early warning platform

[0096] The data analysis and early warning platform is used to obtain the required data from the data center according to the needs and analyze the data, and feed back the analysis results and early warning information to the data center for storage, and send the early warning information to relevant managers; it includes:

[0097] (1) Advanced Data Analysis Module

[0098] ① Environmental change model building unit: It uses supervised learning and unsupervised learning algorithms, such as random forests, support vector machines, time series analysis, etc., to conduct in-depth mining of historical data and build environmental change models to solve a series of scientific, management and policy issues related to the environment, such as climate change research, ecosystem management, pollution control and governance, sustainable use of resources, urban planning and land use, disaster risk management, agriculture and food safety;

[0099] The specific steps of constructing the environmental change model are:

[0100] Step 1: Collect historical and current environmental data according to the type of environmental change to be monitored (natural change and man-made change), and pre-process the data, wherein the current environmental data is obtained from the environmental monitoring data sent by the data transmission module, and the historical data is obtained from environmental monitoring stations, government agencies and research institutions, international organizations, public databases, and scientific literature;

[0101] Step 2: Select the appropriate model type in the advanced data analysis module according to actual needs, such as statistical model (regression analysis, time series analysis), machine learning model (random forest, neural network), physical model or hybrid model, set model parameters and variable factors, and substitute the preprocessed data in step 1 to continuously train and optimize the model;

[0102] Step 3: Use the optimized model to simulate future environmental changes, generate prediction results based on different scenarios (such as climate change scenarios), and use Monte Carlo simulation, Bayesian statistics, or criticality analysis methods in existing technologies to perform uncertainty analysis, and visualize the final results to generate a rate of change or probability map or dynamic simulation video;

[0103] Step 4: Compare the predicted results with the actual situation or subsequent observation data, evaluate the model, and adjust the model parameters according to the evaluation results until the model meets the preset error threshold and outputs the constructed environmental change model.

[0104] ② Anomaly detection unit: Based on historical and real-time monitored environmental data, statistical methods are first used to quickly screen out obvious outliers, and then deep learning algorithms are used to identify abnormal fluctuations in environmental parameters, such as sudden changes in air quality and abnormal increases in noise.

[0105] The anomaly detection unit can feed back the detected anomaly data to the environmental change model to update model parameters or adjust the prediction strategy. The environmental change model can provide prediction data to help the anomaly detection unit better understand data trends and assist in anomaly detection.

[0106] Among them, the deep learning algorithm uses a long short-term memory network (Long Short-Term Memory, LSTM). The LSTM model is trained to predict future time series values, and then the difference between the actual value and the predicted value is compared. A large difference may indicate anomalies.

[0107] The statistical methods include the following two:

[0108] 1) Box plot analysis (interquartile range IQR method)

[0109] Sorting: Sort all observations from smallest to largest.

[0110] Calculate the quartiles: First quartile (Q1): The upper limit of the smallest 25% of all values ​​in the data. Third quartile (Q3): The upper limit of the smallest 75% of all values ​​in the data.

[0111] Calculate the interquartile range (IQR): IQR = Q3 - Q1.

[0112] Define boundaries: Lower boundary: Q1-1.5*IQR. Upper boundary: Q3+1.5*IQR.

[0113] Identify outliers: Any value less than the lower boundary or greater than the upper boundary is considered an outlier.

[0114] 2) Z-score method

[0115] Calculate mean (μ): Calculate the average of all observations.

[0116] Calculate the standard deviation (σ): Calculate the average of the sum of the squares of the differences between all observations and the mean, and then take the square root to get the standard deviation.

[0117] Calculate the Z-score: For each observation x, calculate its Z-score, which is (x-μ) / σ.

[0118] Set the threshold: Usually, if the Z-score of an observation is greater than 3 or less than -3, it is considered an outlier (this threshold can be adjusted according to actual conditions):

[0119] Formula: Z = (x-μ) / σ.

[0120] Specifically, the specific steps for using statistical methods to quickly screen out obvious outliers are:

[0121] Step 1: Obtain historical and real-time monitoring environmental data;

[0122] Collect historical and real-time environmental data from environmental monitoring stations, sensor networks, satellite remote sensing images, weather stations, etc. These environmental data include but are not limited to the concentration of major pollutants in the air (such as PM2.5, PM10, SO2, NOx), temperature, humidity, wind speed, wind direction, air pressure changes, etc.;

[0123] Step 2: Perform statistical analysis on the data through box plot analysis to obtain outliers;

[0124] Step 2.1.1: Draw a box plot. For each environmental parameter (such as PM2.5, temperature, etc.), draw a box plot separately;

[0125] Step 2.1.2: Determine the quartiles, calculate the first quartile (Q1), the third quartile (Q3), and the interquartile range (IQR = Q3-Q1);

[0126] Step 2.1.3: Identify outliers. According to the box plot rule, any data point smaller than Q1-1.5*IQR or larger than Q3+1.5*IQR is considered an outlier.

[0127] Step 2.1.4: Record and mark any outliers, which may be due to equipment failure, extreme weather events, or other unusual circumstances.

[0128] Step 3: Based on the results obtained in step 2, the environmental data are analyzed using the Z-score method to obtain standardized data;

[0129] Step 3.1.1: Calculate the mean and standard deviation: For each environmental parameter, calculate the mean (μ) and standard deviation (σ) of its historical data;

[0130] Step 3.1.2: Record data points with an absolute Z-score greater than 3 as outliers (i.e., data points that are more than 3 standard deviations away from the mean);

[0131] Step 3.1.3: Convert all data points to Z-score to obtain a standardized data set.

[0132] Step 4: Clean the data according to the results of step 3 to obtain a clean data set;

[0133] Step 4.1.1: Handling outliers;

[0134] Specifically, outliers are deleted or interpolation methods (such as linear interpolation, moving average, etc.) are used to correct outliers, and these deleted outliers and their reasons are recorded for reference in subsequent analysis.

[0135] Step 4.1.2: Data smoothing, using sliding window averaging to reduce noise and improve data quality;

[0136] Step 4.1.3: Validate the processed data to ensure that it is statistically reasonable and does not introduce new biases, ultimately outputting a clean, standardized historical and real-time environmental dataset for further analysis, modeling, and decision support.

[0137] (2) Real-time warning and response module

[0138] This module is used to issue an alarm based on the detection results of the abnormality detection unit and send response suggestions to the operating user; it includes:

[0139] ①Threshold setting unit

[0140] Used to set alert and warning thresholds for various environmental indicators in accordance with national environmental standards and local regulations;

[0141] ② Early warning unit

[0142] When an abnormal event is received from the abnormality detection unit, it is used to send an alarm message to the operating user via SMS, email or APP push;

[0143] ③Response suggestion unit

[0144] Used to generate preliminary emergency response recommendations based on alarm information, and assist decision makers in quickly formulating response measures.

[0145] 5. User side

[0146] The user end includes a web end and a mobile APP, which is used to provide users with a web page and a mobile-friendly visualization interface, visually displaying the distribution of monitoring points, real-time environmental data query, historical data analysis and comparison, and providing an environmental event reporting interface for the public to report environmental events such as illegal discharge of pollutants, abnormal odors, etc. through the APP.

[0147] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that they can still modify or replace the technical solution of the present invention with equivalents, and these modifications or equivalent replacements cannot cause the modified technical solution to deviate from the spirit and scope of the technical solution of the present invention.

Claims

1. A high-precision regional environment monitoring system based on a communication tower group, deployed on the communication towers in the area to be monitored, characterized in that: include: Environmental monitoring module, used to monitor environmental parameter data in real time through multiple sensors; A data transmission module, used to transmit the monitored environmental parameter data between multiple communication towers in the area and transmit the environmental parameter data to a data center; A data center, used to collect and store environmental monitoring data transmitted by the data transmission module; The data analysis and early warning platform is used to obtain the corresponding data from the data center for data analysis according to actual needs, and feed back the analysis results and early warning information to the data center for storage, and send the early warning information to relevant managers; The user side, including the web side and mobile APP, is used to provide users with web pages and mobile-friendly visualization interfaces, visually display the distribution of monitoring points, real-time environmental data query, historical data analysis and comparison, and provide an environmental event reporting interface for the public to report environmental events through the APP.

2. A high-precision regional environment monitoring system based on a communication tower group as claimed in claim 1, characterized in that: The environmental monitoring module comprises: An air quality monitoring unit, used for monitoring the concentration of main pollutants in the air in real time through a first sensor group; A meteorological parameter monitoring unit, used for real-time monitoring of temperature, humidity, wind speed and direction, and air pressure through a second sensor group; A noise pollution monitoring unit is used to collect current environmental noise data through acoustic and electrical sensors and obtain the level of noise pollution according to noise standards; Special environmental factor monitoring unit, used to monitor radiation dose, water quality, air pollution, soil quality, noise level, biodiversity and some specific pollutants through the optional third sensor group.

3. A high-precision regional environment monitoring system based on a communication tower group as claimed in claim 2, characterized in that: The first sensor group includes a PM2.5 / PM10 particulate matter sensor, a sulfur dioxide gas sensor, a nitrogen dioxide gas sensor, a carbon monoxide gas sensor and an ozone gas sensor; The second sensor group includes a humidity sensor, an anemometer, and a barometer; The third sensor group includes a radiation dosimeter and a water quality monitoring probe.

4. A high-precision regional environment monitoring system based on a communication tower group as claimed in claim 3, characterized in that: The data transmission module comprises: A wireless transmission unit, used to transmit the environmental parameter data monitored by the environmental monitoring module to the data center via a low-power wide area network; The data encryption unit is used to perform AES encryption transmission on the original environment parameter data.

5. A high-precision regional environment monitoring system based on a communication tower group as claimed in claim 4, characterized in that: The data center includes: Distributed storage database for storing and retrieving collected data; The elastic computing module is used to utilize the elastic computing service provided by the cloud service provider to determine the normal and peak load data of the monitoring system, wherein the load data includes CPU usage, memory requirements, storage space and network bandwidth.

6. A high-precision regional environment monitoring system based on a communication tower group as claimed in claim 5, characterized in that: The data analysis and early warning platform includes: Advanced data analysis module for identifying abnormal fluctuations in environmental parameters based on current environmental data and historical data; The real-time warning and response module is used to issue an alarm based on the detection results of the anomaly detection unit and send response suggestions to the operating user.

7. A high-precision regional environment monitoring system based on a communication tower group as claimed in claim 6, characterized in that: The advanced data analysis module includes: An environment change model building unit is used to build an environment change model, provide prediction data to the anomaly detection unit through the environment change model, and assist the anomaly detection unit in analyzing data trends; The anomaly detection unit, based on historical and real-time monitored environmental data, first uses statistical methods to quickly screen out obvious outliers, then uses deep learning algorithms to identify abnormal fluctuations in environmental parameters, and feeds the abnormal data back to the environmental change model to update the model parameters.

8. A high-precision regional environment monitoring system based on a communication tower group as claimed in claim 7, characterized in that: The real-time warning and response module includes: Threshold setting unit, used to set the alert and warning thresholds of various environmental indicators according to national environmental standards and local regulations; An early warning unit is used to send warning information to the operating user; The response suggestion unit is used to generate preliminary emergency response suggestions based on the type of abnormal event.

9. A high-precision regional environment monitoring system based on a communication tower group as claimed in claim 8, characterized in that: The steps of constructing the environmental change model include: Step 1: Collect historical and current environmental data according to the type of environmental change to be monitored, and pre-process the collected environmental data; Step 2: Select the appropriate model type in the advanced data analysis module according to actual needs, set model parameters and variable factors, and input the data preprocessed in step 1 into the selected model to train and optimize the model; Step 3: Use the optimized model to simulate future environmental changes, generate prediction results based on different scenarios, and perform uncertainty analysis; Step 4: Compare the predicted results with the actual situation or subsequent observation data, evaluate the model, and adjust the model parameters according to the evaluation results until the model meets the preset error threshold and outputs the constructed environmental change model.

10. A high-precision regional environment monitoring system based on a communication tower group as claimed in claim 9, characterized in that: The specific steps of the anomaly detection unit using statistical methods to screen outliers of environmental parameters include: Step 1: Obtain historical and real-time monitoring environmental data; Step 2: Perform statistical analysis on the data through box plot analysis to obtain outliers; Step 2.1.1: Draw a box plot. For each environmental parameter, draw a box plot separately; Step 2.1.2: Determine the quartiles, calculate the first quartile Q1, the third quartile Q3, and the interquartile range IQR = Q3-Q1; Step 2.1.3: According to the box plot rules, any data point less than Q1-1.5*IQR or greater than Q3+1.5*IQR is considered an outlier; Step 2.1.4: Record and mark all outliers; Step 3: Based on the results obtained in step 2, the environmental data are analyzed using the Z-score method to obtain standardized data; Step 3.1.1: Calculate the mean and standard deviation: For each environmental parameter, calculate the mean μ and standard deviation σ of its historical data; Step 3.1.2: Record the data points with Z-score absolute value greater than 3 as outliers; Step 3.1.3: Convert all data points to Z-score to obtain a standardized data set; Step 4: Clean the standardized data to obtain a clean environmental data set; Step 4.1.1: Handling outliers; Specifically, outliers are deleted or corrected using interpolation methods, and all deleted outliers and their causes are recorded; Step 4.1.2: Use sliding window averaging to reduce noise; Step 4.1.3: Verify the processed data and output the final historical and real-time environmental datasets.

Images