Environment monitoring alarm method and platform based on mobile communication

Abstract

The invention discloses an environment monitoring alarm method and platform based on mobile communication, and relates to the technical field of environment monitoring, and the method comprises the steps: constructing a mobile communication network between environment monitoring points; analyzing and quantifying dynamic association attributes among the environment monitoring points; screening out an associated neighbor environment monitoring point set based on each environment monitoring point; calculating and constructing real-time derivation data of an environment verification monitoring point corresponding to the environment monitoring point; acquiring actual monitoring data of the environment monitoring point in real time through a mobile communication network; and comparing the actual monitoring data with the real-time derivation data, and if a comparison result exceeds a preset threshold value, generating an alarm signal and sending the alarm signal through a mobile communication network. The method has the advantages that the dynamic association attribute between the environment monitoring points is quantified by analyzing the historical data, the perception capability of real environment abnormal events is enhanced, the intelligent operation and maintenance level is improved, and the operation and maintenance cost is reduced.

Description

Technical Field

[0001] This invention relates to the field of environmental monitoring technology, specifically to an environmental monitoring and alarm method and platform based on mobile communication. Background Technology

[0002] Currently, environmental monitoring systems are widely used in air quality, water quality, and pollution source monitoring. These systems typically rely on multiple monitoring points distributed across a vast area, using mobile communication networks for remote data transmission and aggregation. However, most existing systems only perform independent threshold checks on data from each monitoring point, triggering an alarm whenever data from a particular point becomes abnormal. This method has significant limitations: firstly, it is susceptible to false alarms due to temporary equipment failures, transmission interference, or transient environmental disturbances, such as vehicle exhaust, leading to low alarm reliability and wasted maintenance resources; secondly, it lacks utilization of the inherent correlations between monitoring points, making it impossible to effectively cross-validate data from surrounding points when data from a single monitoring point is abnormal, thus making it difficult to distinguish between a real environmental event and an equipment malfunction.

[0003] Furthermore, traditional monitoring methods assess each monitoring point in isolation, failing to fully explore and utilize the spatial and temporal correlations inherent in the monitoring network itself. When an anomaly occurs at a monitoring point, maintenance personnel often need to manually retrieve data from surrounding sites for comparison and analysis, resulting in slow response times, reliance on manual experience, and low efficiency. Therefore, the industry urgently needs a method that can intelligently perceive the dynamic correlations between monitoring points and utilize these correlations to perform real-time, automated verification of data from individual monitoring points, thereby significantly improving the accuracy and reliability of alarms and reducing maintenance costs. Summary of the Invention

[0004] To address the aforementioned technical problems, this paper provides an environmental monitoring and alarm method and platform based on mobile communication. This technical solution solves at least one of the technical problems mentioned in the background section.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows: An environmental monitoring and alarm method based on mobile communication includes: Construct a mobile communication network between environmental monitoring sites; Historical data from multiple environmental monitoring points are acquired, and the dynamic correlation attributes between the environmental monitoring points are analyzed and quantified based on the historical data. Based on each environmental monitoring point, a set of related neighbor environmental monitoring points is selected from its related environmental monitoring points according to the dynamic association attributes. Based on the real-time monitoring data of each associated environmental monitoring point in the set of associated neighbor environmental monitoring points and their corresponding dynamic association attributes, real-time derivation data of the environmental verification monitoring point corresponding to the environmental monitoring point is constructed by calculation. The actual monitoring data of the environmental monitoring points are obtained in real time through the mobile communication network; The actual monitoring data is compared with the real-time derived data. If the comparison result exceeds a preset threshold, an alarm signal is generated and sent through the mobile communication network.

[0006] Preferably, the step of acquiring historical data from multiple environmental monitoring points, and analyzing and quantifying the dynamic correlation attributes between the environmental monitoring points based on the historical data, specifically includes: Calculate at least one of the following: spatial distance weight, Pearson correlation coefficient, or dynamic time warping (DTW) similarity between any two historical data sequences of environmental collection points; The calculation results are normalized to obtain the correlation strength value between any two environmental collection points, and a correlation strength value matrix is ​​formed.

[0007] Preferably, for each environmental monitoring point, selecting a set of associated neighbor environmental monitoring points from its associated environmental monitoring points based on the dynamic association attribute specifically includes: For each environmental monitoring point, extract its correlation strength value with all other environmental collection points from the correlation weight matrix; Select the associated environmental monitoring points whose association strength value is greater than the preset association threshold or whose association strength value ranks among the top K, and form the associated neighbor set of that environmental monitoring point.

[0008] Preferably, the step of constructing real-time derivation data of the environmental verification monitoring point corresponding to the environmental monitoring point by calculating the real-time monitoring data of each associated environmental monitoring point in the associated neighbor environmental monitoring point set and its corresponding dynamic association attributes specifically includes: The environmental monitoring points used for the derivation are recorded as the target environmental monitoring points; Based on the monitoring data of each associated environmental monitoring point of the target environmental monitoring point, analyze the monitoring prediction value of each associated environmental monitoring point for the target environmental monitoring point; Based on the association strength value of each associated environmental monitoring point in the associated neighbor set, the monitoring prediction values ​​of all associated environmental monitoring points for the target environmental monitoring point are comprehensively fitted to obtain the real-time derivation data of the environmental verification monitoring point corresponding to the environmental monitoring point.

[0009] Preferably, the step of analyzing the predicted monitoring value of each associated environmental monitoring point for the target environmental monitoring point based on the monitoring data of each associated environmental monitoring point of the target environmental monitoring point specifically includes: The monitoring data of the target environmental monitoring point and the monitoring data of the related environmental monitoring point are matched according to the time sequence, and several sample data pairs are selected. The sample data pairs cover the fluctuation range of the monitoring data of the target environmental monitoring point and the fluctuation range of the monitoring data of the related environmental monitoring point. Based on sample data, regression equation fitting or similar data fitting is used to analyze the monitoring prediction values ​​of the target environmental monitoring points by the associated environmental monitoring points.

[0010] Preferably, the regression equation fitting specifically includes: A linear regression equation was constructed using the monitoring data of the associated environmental monitoring points in the sample data pair as the independent variable and the monitoring data of the target environmental monitoring point in the sample data pair as the dependent variable. By substituting the real-time monitoring data of the associated environmental monitoring points into the linear regression equation, the predicted monitoring values ​​of the associated environmental monitoring points for the target environmental monitoring point are obtained.

[0011] Preferably, the similarity data fitting specifically includes: In the sample data pairs, the sample data pairs whose monitoring data of the associated environmental monitoring point is greater than and closest to the real-time monitoring data of the associated environmental monitoring point are used as the upper similar data; In the sample data pairs, the sample data pairs whose monitoring data of the associated environmental monitoring point is less than and most similar to the real-time monitoring data of the associated environmental monitoring point are selected as the next similar data; Based on the proportion of real-time monitoring data from associated environmental monitoring points in the upper and lower similar data, the predicted monitoring value of associated environmental monitoring points for the target environmental monitoring point is calculated.

[0012] Furthermore, a mobile communication-based environmental monitoring and alarm platform is proposed to implement the mobile communication-based environmental monitoring and alarm method described above, including: The mobile communication network construction module is used to build a mobile communication network between environmental monitoring points; The historical data acquisition module is used to acquire historical data from multiple environmental monitoring points; The dynamic correlation analysis module is used to analyze and quantify the dynamic correlation attributes between various environmental monitoring points based on the historical data. The associated neighbor filtering module is used to filter out a set of associated neighbor environmental monitoring points from each associated environmental monitoring point based on the dynamic association attributes. The real-time derivation data construction module is used to construct real-time derivation data of the environmental verification monitoring point corresponding to the environmental monitoring point by calculating based on the real-time monitoring data of each associated environmental monitoring point in the set of associated neighbor environmental monitoring points and their corresponding dynamic association attributes. The actual data acquisition module is used to acquire the actual monitoring data of the environmental monitoring point in real time through a mobile communication network; The comparison alarm module is used to compare the actual monitoring data with the real-time derived data. If the comparison result exceeds a preset threshold, an alarm signal is generated and sent through the mobile communication network.

[0013] Optionally, the dynamic correlation analysis module specifically includes: The correlation strength calculation unit is used to calculate at least one of the following: spatial distance weight, Pearson correlation coefficient, or dynamic time warping (DTW) similarity between any two historical data sequences of environmental collection points. The normalization processing unit is used to normalize the calculation results to obtain the correlation strength value between any two environmental collection points and form a correlation strength value matrix.

[0014] Optionally, the real-time inference data construction module specifically includes: The prediction value analysis unit is used to analyze the predicted monitoring value of each associated environmental monitoring point for the target environmental monitoring point based on the monitoring data of each associated environmental monitoring point of the target environmental monitoring point. The comprehensive fitting unit is used to comprehensively fit the monitoring prediction values ​​of all associated environmental monitoring points for the target environmental monitoring point based on the association strength value of each associated environmental monitoring point in the associated neighbor set, so as to obtain the real-time derivation data of the environmental verification monitoring point corresponding to the environmental monitoring point.

[0015] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention quantifies the dynamic correlation attributes between environmental monitoring points by analyzing historical data, and constructs theoretical values ​​for target points using real-time data from neighboring monitoring points. These theoretical values ​​are then intelligently compared with actual measured values, achieving cross-validation and self-verification of single-point monitoring data. This method effectively identifies isolated data anomalies caused by equipment failure or transient interference, significantly reducing the system's false alarm rate and improving the accuracy and reliability of alarm information. Simultaneously, the system enhances its ability to perceive real-world environmental anomalies, improves system robustness and intelligent operation and maintenance, reduces reliance on manual verification, and lowers operation and maintenance costs. Attached Figure Description

[0016] Figure 1 This is a flowchart of the environmental monitoring and alarm method based on mobile communication proposed in this invention; Figure 2 The flowchart of the method for analyzing and quantifying the dynamic correlation attributes between various environmental monitoring points proposed in this invention is shown below. Figure 3 This is a flowchart of the method for filtering out a set of related neighbor environmental monitoring points proposed in this invention; Figure 4 This is a flowchart of the method for constructing real-time derivation data of environmental verification monitoring points corresponding to environmental monitoring points, as proposed in this invention. Detailed Implementation

[0017] The following description is intended to disclose the invention and enable those skilled in the art to implement it. The preferred embodiments described below are merely examples, and other obvious variations will occur to those skilled in the art.

[0018] Example 1: Reference Figure 1 As shown, this embodiment proposes an environmental monitoring and alarm method based on mobile communication, including: Constructing a mobile communication network between environmental monitoring points, the coverage advantage of mobile communication networks (such as 4G / 5G / NB-IoT) makes it possible to flexibly deploy monitoring points in various complex geographical environments, solving the problems of difficult and costly traditional wired network deployment, and ensuring that monitoring data and alarm signals can be stably and efficiently transmitted to the monitoring center or cloud platform. Historical data from multiple environmental monitoring points are acquired, and the dynamic correlation attributes between these points are analyzed and quantified based on the historical data. By analyzing long-term historical data, the degree of mutual influence of environmental parameters between any two points is objectively measured, thereby establishing a quantitative dynamic correlation model that can be adjusted as data is updated over time, providing a basis for subsequent intelligent verification. Based on each environmental monitoring point, a set of neighboring environmental monitoring points is selected from its associated environmental monitoring points according to dynamic correlation attributes. Not all monitoring points are strongly correlated with it. By setting a correlation threshold or selecting the Top-K strongest correlated points, the set of neighboring points that best reflects the normal change pattern of that point can be selected. This operation optimizes computing resources and ensures that the information used for subsequent inferences comes from highly reliable correlation sources. Based on the real-time monitoring data of each associated environmental monitoring point in the set of associated neighbor environmental monitoring points and their corresponding dynamic association attributes, real-time inferred data of the environmental verification monitoring points corresponding to the environmental monitoring points are calculated and constructed. Using the real-time data of the selected "associated neighbors" and comprehensively considering their association strength with the target point, a theoretical expected value of the target point at the current moment is simulated through data fitting algorithms such as weighted averaging and regression prediction. This inferred value represents the reasonable range that the target point should be within under normal circumstances. By utilizing mobile communication networks, real-time monitoring data from environmental monitoring points is acquired. This actual monitoring data serves as the direct basis for alarm judgment, and its real-time nature ensures that the system can respond promptly to environmental changes. Mobile communication networks guarantee that this real-time data can be collected with low latency and compared with derived data at the same time reference. The actual monitoring data is compared with the real-time derived data. If the comparison result exceeds a preset threshold, an alarm signal is generated and sent via the mobile communication network. When the actual measured value deviates significantly from the theoretical value calculated based on the associated neighbor data, it indicates that the anomaly is likely caused by equipment failure at the target point itself or a local real event, rather than a normal regional environmental fluctuation, thus triggering a high-confidence alarm. This method greatly reduces false alarms caused by drift of a single sensor or transient interference.

[0019] Example 2: Reference Figure 2 As shown, based on Embodiment 1, this embodiment further proposes to acquire historical data from multiple environmental monitoring points, and to analyze and quantify the dynamic correlation attributes between these monitoring points based on the historical data. Specifically, this includes: Calculate at least one of the following: spatial distance weight, Pearson correlation coefficient, or dynamic time warping (DTW) similarity between any two historical data sequences of environmental collection points; The specific calculation method for spatial distance weight is as follows: ; In the formula, The spatial distance weight between the i-th environmental data collection point and the j-th environmental data collection point is... Let be the physical distance between the i-th environmental sampling point and the j-th environmental sampling point.

[0020] The Pearson correlation coefficient is calculated as follows: Align the historical data from the two environmental data collection points; Calculate and evaluate the Pearson correlation coefficient r of historical data from two environmental data collection points; ; in, , These are historical data from two environmental data collection points. , These are the historical data averages from two environmental data collection points. The number of data point pairs analyzed; The closer the value is to 1, the stronger the real-time correlation between the historical data of the two environmental collection points.

[0021] The method for evaluating DTW similarity is as follows: Z-score normalization was used to normalize the historical data from the two environmental data collection points. Calculate the Euclidean distance between all point pairs in the historical data of the two environmental acquisition points; Find a path with the minimum cumulative distance. The cumulative distance of the path is used to evaluate the DTW similarity of historical data from two environmental collection points.

[0022] The calculation results are normalized to obtain the correlation strength value between any two environmental collection points, and a correlation strength value matrix is ​​formed.

[0023] This method utilizes three complementary mathematical tools—spatial distance weighting, Pearson correlation coefficient, and Dynamic Time Warping (DTW) similarity—to comprehensively assess the strength of point-to-point associations from different dimensions: physical proximity, linear correlation, and temporal pattern similarity. Spatial distance weighting reflects the physical law that environmental factors decay with geographical distance; the Pearson correlation coefficient captures the synchronicity and linear correlation of changes in monitoring point data; and DTW similarity effectively characterizes complex associations that may have phase differences in time but similar change patterns. The calculation results of these different dimensions are normalized and integrated into an association strength value matrix, transforming the complex spatial topology and temporal relationships into a quantitative relationship model that can be directly processed and analyzed by a computer. This lays a precise and reliable mathematical foundation for subsequent intelligent screening of "related neighbors" and the construction of highly reliable real-time inference data. This method overcomes the limitations of relying solely on physical distance to determine association, enabling the system to adaptively capture and utilize the dynamic data patterns that truly exist in the monitoring network.

[0024] Specifically, refer to Figure 3 As shown, this embodiment also proposes to select a set of related neighboring environmental monitoring points from its related environmental monitoring points based on dynamic association attributes for each environmental monitoring point. Specifically, this includes: For each environmental monitoring point, extract its correlation strength value with all other environmental collection points from the correlation weight matrix; Select the associated environmental monitoring points whose association strength value is greater than the preset association threshold or whose association strength value ranks among the top K, and form the associated neighbor set of that environmental monitoring point.

[0025] The system achieves intelligent optimization and adaptive construction of the monitoring network structure. By dynamically selecting the most relevant neighbor set with the highest correlation strength or meeting the threshold for each monitoring point, the system can accurately focus on the most valuable data source, effectively removing noise interference from weakly correlated or irrelevant monitoring points in subsequent data derivation and verification. This not only significantly reduces computational complexity and improves system real-time performance, but more importantly, it ensures that the real-time derivation data constructed for each target point is based on the most reliable and relevant correlation information, thereby greatly improving the accuracy of anomaly detection and the credibility of alarms, making the system alarms truly targeted and valuable.

[0026] Example 3: Reference Figure 4As shown, this embodiment, based on embodiment two, further proposes to construct real-time derived data of environmental verification monitoring points corresponding to environmental monitoring points by calculating real-time monitoring data of each associated environmental monitoring point in the associated neighbor environmental monitoring point set and its corresponding dynamic association attributes. Specifically, this includes: The environmental monitoring points used for the derivation are recorded as the target environmental monitoring points; Based on the monitoring data of each associated environmental monitoring point of the target environmental monitoring point, analyze the monitoring prediction value of each associated environmental monitoring point for the target environmental monitoring point; Based on the association strength value of each associated environmental monitoring point in the associated neighbor set, the monitoring prediction values ​​of all associated environmental monitoring points for the target environmental monitoring point are comprehensively fitted to obtain the real-time derivation data of the environmental verification monitoring point corresponding to the environmental monitoring point.

[0027] Specifically, the formula for calculating the real-time derived data from environmental verification monitoring points is as follows: ; In the formula, This is the real-time derived data for the k-th environmental verification monitoring point. Let k be the set of associated neighbors of the k-th environmental verification monitoring point. For the j-th associated environmental monitoring point in the associated neighbor set, Let be the association strength value between the j-th associated environmental monitoring point and the k-th environmental verification monitoring point in the associated neighbor set. This represents the predicted monitoring value of the j-th associated environmental monitoring point in the associated neighbor set for the k-th environmental verification monitoring point.

[0028] Significant benefits have been achieved by introducing a real-time inference data calculation formula based on association strength weighting. This formula innovatively combines real-time monitoring data of associated neighbors with their association weights for weighted comprehensive fitting. This approach not only fully utilizes the real-time information of all highly associated neighbor points but also distinguishes the credibility of different neighbor point contribution values ​​through weight differences, making the final generated real-time inference data more accurate and reliable. Compared to simple averaging or methods relying on a single neighbor point, this weighted model more scientifically reflects the expected state of the target point under normal conditions, greatly reducing the sensitivity to instantaneous fluctuations in individual neighbor point data. This provides a dynamic, robust, and highly reliable reference benchmark for subsequent anomaly comparisons, fundamentally improving the accuracy of alarm decisions.

[0029] When comparing actual monitoring data with real-time derived data, the deviation rate is calculated as the ratio of the absolute difference between the actual monitoring data and the real-time derived data to the actual monitoring data. The deviation rate is then compared with a preset threshold. The preset threshold is dynamically set based on the sum of the association strength values ​​of all elements in the associated neighbor set. The larger the association strength value of all elements in the associated neighbor set, the more accurate the real-time derived data is, and the preset threshold decreases. Conversely, the smaller the association strength value of all elements in the associated neighbor set, the less accurate the real-time derived data is, and the preset threshold increases.

[0030] Specifically, based on the monitoring data of each associated environmental monitoring point of the target environmental monitoring point, the analysis of the predicted monitoring value of each associated environmental monitoring point for the target environmental monitoring point includes: The monitoring data of the target environmental monitoring point and the monitoring data of the related environmental monitoring point are matched according to the time sequence, and several sample data pairs are selected. The sample data pairs cover the fluctuation range of the monitoring data of the target environmental monitoring point and the fluctuation range of the monitoring data of the related environmental monitoring point. Based on sample data, regression equation fitting or similar data fitting is used to analyze the monitoring prediction values ​​of the target environmental monitoring points by the associated environmental monitoring points.

[0031] Specifically, for some environmental monitoring points with obvious linear relationships in their monitoring data, regression equations are used for fitting, including: A linear regression equation was constructed using the monitoring data of the associated environmental monitoring points in the sample data pair as the independent variable and the monitoring data of the target environmental monitoring point in the sample data pair as the dependent variable. By substituting the real-time monitoring data of the associated environmental monitoring points into the linear regression equation, the predicted monitoring values ​​of the associated environmental monitoring points for the target environmental monitoring point are obtained.

[0032] The use of regression equation fitting to obtain monitoring prediction values ​​has yielded significant benefits. When a clear linear relationship exists between the monitoring data of associated environmental monitoring points and target points, the linear regression equation constructed based on historical data can accurately characterize the stable and quantifiable change patterns between the two using a simple mathematical model. This method is computationally efficient and has a fast response speed, making it highly suitable for monitoring scenarios with stringent real-time requirements. Substituting real-time data from associated points into the equation quickly yields scientific predictions of the target points, greatly reducing reliance on complex models and massive computing resources, and ensuring efficient system operation. Furthermore, the regression relationship established based on historical data has strong statistical significance, making the prediction results more objective and reliable. This provides a solid theoretical basis for subsequent data comparison and anomaly detection, effectively improving the accuracy and practicality of the entire alarm system.

[0033] Specifically, for some environmental monitoring points where the linear relationship between monitoring data is not obvious, similar data fitting is used, including: In the sample data pairs, the sample data pairs whose monitoring data of the associated environmental monitoring point is greater than and closest to the real-time monitoring data of the associated environmental monitoring point are used as the upper similar data; In the sample data pairs, the sample data pairs whose monitoring data of the associated environmental monitoring point is less than and most similar to the real-time monitoring data of the associated environmental monitoring point are selected as the next similar data; Based on the proportion of real-time monitoring data from associated environmental monitoring points in the upper and lower similar data, the predicted monitoring value of associated environmental monitoring points for the target environmental monitoring point is calculated.

[0034] The specific calculation formula is as follows: ; To correlate the predicted monitoring values ​​of the target environmental monitoring point with the environmental monitoring points, These are the monitoring data values ​​of the target environmental monitoring points in the similar data above. The values ​​are the monitoring data of the target environmental monitoring points in the similar data set below. These are the monitoring data values ​​of the associated environmental monitoring points in the similar data above. The following are the monitoring data values ​​of related environmental monitoring points in similar data. This refers to real-time monitoring data from associated environmental monitoring points; When the linear relationship between monitoring data is not obvious, a linear simulation method between similar data is used to fit the predicted monitoring values ​​of the associated environmental monitoring points for the target environmental monitoring points. This cleverly avoids the difficulty of constructing complex nonlinear models, instead employing a local linear approximation strategy. It retrieves the two historical sample points closest to the current state in real time and performs interpolation between them. This approach has strong adaptability and fault tolerance for nonlinear and highly volatile data relationships. It does not rely on a holistic, fixed function form, but dynamically predicts based on the local position of the current data point in the historical distribution, thus capturing complex relationships between data more flexibly and accurately. This not only ensures the rationality of the predicted values ​​under nonlinear relationships, greatly expanding the applicability of this method, but also significantly improves the robustness and predictive reliability of the system when facing complex environmental data.

[0035] Example 4: This embodiment proposes an environmental monitoring and alarm platform based on mobile communication, including: The mobile communication network construction module is used to build a mobile communication network between environmental monitoring points; The historical data acquisition module is used to acquire historical data from multiple environmental monitoring points; The dynamic correlation analysis module is used to analyze and quantify the dynamic correlation attributes between various environmental monitoring points based on historical data. The associated neighbor filtering module is used to filter out a set of associated neighbor environmental monitoring points from each environmental monitoring point based on its dynamic association attributes. The real-time derivation data construction module is used to construct real-time derivation data of the environmental verification monitoring point corresponding to the environmental monitoring point by calculating based on the real-time monitoring data of each associated environmental monitoring point in the set of associated neighbor environmental monitoring points and their corresponding dynamic association attributes. The actual data acquisition module is used to acquire actual monitoring data from environmental monitoring points in real time via mobile communication networks; The comparison alarm module is used to compare the actual monitoring data with the real-time derived data. If the comparison result exceeds the preset threshold, an alarm signal is generated and sent through the mobile communication network.

[0036] In some preferred embodiments, the dynamic correlation analysis module specifically includes: The correlation strength calculation unit is used to calculate at least one of the following: spatial distance weight, Pearson correlation coefficient, or dynamic time warping (DTW) similarity between any two historical data sequences of environmental collection points. The normalization processing unit is used to normalize the calculation results to obtain the correlation strength value between any two environmental collection points and form a correlation strength value matrix.

[0037] In some preferred embodiments, the real-time inference data construction module specifically includes: The prediction value analysis unit is used to analyze the predicted monitoring value of each associated environmental monitoring point for the target environmental monitoring point based on the monitoring data of each associated environmental monitoring point of the target environmental monitoring point. The comprehensive fitting unit is used to comprehensively fit the monitoring prediction values ​​of all associated environmental monitoring points to the target environmental monitoring point based on the association strength value of each associated environmental monitoring point in the associated neighbor set, so as to obtain the real-time derivation data of the environmental verification monitoring point corresponding to the environmental monitoring point.

[0038] In summary, the advantages of this invention are as follows: by analyzing historical data, it quantifies the dynamic correlation attributes between environmental monitoring points, and uses real-time data from neighboring monitoring points to construct theoretical values ​​for target points. These theoretical values ​​are then intelligently compared with actual measured values, thereby achieving cross-validation and self-verification of single-point monitoring data. This method can effectively identify isolated data anomalies caused by equipment failure or transient interference, significantly reducing the system's false alarm rate and improving the accuracy and reliability of alarm information. Simultaneously, the system enhances its ability to perceive real-world environmental anomalies, improves system robustness and intelligent operation and maintenance, reduces reliance on manual verification, and lowers operation and maintenance costs.

[0039] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the claimed invention. The scope of protection claimed by the appended claims and their equivalents is defined.

Claims

1. A mobile communication-based environmental monitoring alarm method, characterized by, The method comprises the following steps: constructing a mobile communication network between environment monitoring points; acquiring historical data of multiple environment monitoring points, analyzing and quantifying dynamic correlation attributes between the environment monitoring points based on the historical data; based on each environment monitoring point, filtering a set of associated neighbor environment monitoring points from the associated environment monitoring points according to the dynamic correlation attributes; based on the real-time monitoring data of each associated environment monitoring point in the set of associated neighbor environment monitoring points and the corresponding dynamic correlation attributes, constructing real-time derived data of the environment verification monitoring point corresponding to the environment monitoring point through calculation; acquiring actual monitoring data of the environment monitoring point in real time through the mobile communication network; comparing the actual monitoring data with the real-time derived data, and if the comparison result exceeds a preset threshold, generating an alarm signal and sending it through the mobile communication network.

2. The environment monitoring alarm method based on mobile communication according to claim 1, characterized in that, The step of acquiring historical data of multiple environment monitoring points, analyzing and quantifying dynamic correlation attributes between the environment monitoring points based on the historical data specifically comprises: calculating at least one of the spatial distance weight, the Pearson correlation coefficient or the dynamic time warping (DTW) similarity between the historical data sequences of any two environment monitoring points; normalizing the calculation results to obtain the correlation strength value between any two environment monitoring points and form a correlation strength value matrix.

3. The environment monitoring alarm method based on mobile communication according to claim 2, characterized in that, The step of, based on each environment monitoring point, filtering a set of associated neighbor environment monitoring points from the associated environment monitoring points according to the dynamic correlation attributes specifically comprises: for each environment monitoring point, extracting the correlation strength value between it and all other environment monitoring points from the correlation weight matrix; filtering out the associated environment monitoring points with correlation strength values greater than a preset correlation threshold or ranking the top K associated environment monitoring points according to the correlation strength values to form the associated neighbor set of the environment monitoring point.

4. The environment monitoring alarm method based on mobile communication according to claim 3, characterized in that, The step of, based on the real-time monitoring data of each associated environment monitoring point in the set of associated neighbor environment monitoring points and the corresponding dynamic correlation attributes, constructing real-time derived data of the environment verification monitoring point corresponding to the environment monitoring point through calculation specifically comprises: denoting the environment monitoring point to be derived as a target environment monitoring point; analyzing the monitoring prediction value of each associated environment monitoring point for the target environment monitoring point based on the monitoring data of each associated environment monitoring point of the target environment monitoring point; comprehensively fitting the monitoring prediction values of all associated environment monitoring points for the target environment monitoring point based on the correlation strength values of each associated environment monitoring point in the associated neighbor set to obtain the real-time derived data of the environment verification monitoring point corresponding to the environment monitoring point.

5. The environment monitoring alarm method based on mobile communication according to claim 4, characterized in that, The step of analyzing the monitoring prediction value of each associated environment monitoring point for the target environment monitoring point based on the monitoring data of each associated environment monitoring point of the target environment monitoring point specifically comprises: matching the monitoring data of the target environment monitoring point and the monitoring data of the associated environment monitoring point in time sequence, and selecting a plurality of sample data pairs, wherein the sample data pairs cover the fluctuation range of the monitoring data of the target environment monitoring point and the fluctuation range of the monitoring data of the associated environment monitoring point; analyzing the monitoring prediction value of the associated environment monitoring point for the target environment monitoring point based on the sample data pairs by using a regression equation fitting or similar data fitting method.

6. The environment monitoring alarm method based on mobile communication according to claim 5, characterized in that, The regression equation fitting specifically includes: Taking the monitoring data of the associated environmental monitoring point in the sample data pair as the independent variable, and taking the monitoring data of the target environmental monitoring point in the sample data pair as the dependent variable, a linear regression equation is constructed; The real-time monitoring data of the associated environmental monitoring point is substituted into the linear regression equation to obtain the monitoring prediction value of the associated environmental monitoring point for the target environmental monitoring point.

7. The environment monitoring alarm method based on mobile communication according to claim 6, characterized in that, The similar data fitting specifically includes: Retrieving the sample data pair in which the monitoring data of the associated environmental monitoring point is greater than and closest to the real-time monitoring data of the associated environmental monitoring point in the sample data pair as the upper similar data; Retrieving the sample data pair in which the monitoring data of the associated environmental monitoring point is less than and closest to the real-time monitoring data of the associated environmental monitoring point in the sample data pair as the lower similar data; According to the proportion of the real-time monitoring data of the associated environmental monitoring point in the upper similar data and the lower similar data, the monitoring prediction value of the associated environmental monitoring point for the target environmental monitoring point is calculated.

8. A mobile communication based environment monitoring alarm platform, characterized by The mobile communication-based environmental monitoring alarm method according to any one of claims 1-7, comprising: a mobile communication network construction module, configured to construct a mobile communication network between the environmental monitoring points; a historical data acquisition module, configured to acquire historical data of the plurality of environmental monitoring points; a dynamic correlation analysis module, configured to analyze and quantify the dynamic correlation attributes between the environmental monitoring points based on the historical data; an associated neighbor screening module, configured to, based on each environmental monitoring point, screen an associated neighbor environmental monitoring point set from the associated environmental monitoring points of the environmental monitoring point according to the dynamic correlation attributes; a real-time derived data construction module, configured to construct real-time derived data of an environmental verification monitoring point corresponding to the environmental monitoring point by calculation based on the real-time monitoring data of each associated environmental monitoring point in the associated neighbor environmental monitoring point set and the corresponding dynamic correlation attributes; an actual data acquisition module, configured to acquire actual monitoring data of the environmental monitoring point in real time through the mobile communication network; a comparison and alarm module, configured to compare the actual monitoring data with the real-time derived data, and if the comparison result exceeds a preset threshold, generate an alarm signal and send the alarm signal through the mobile communication network.

9. The mobile communication based environment monitoring alarm platform as claimed in claim 8, wherein, The dynamic correlation analysis module specifically includes: a correlation strength calculation unit, configured to calculate at least one of a spatial distance weight, a Pearson correlation coefficient, or a dynamic time warping (DTW) similarity between the historical data sequences of any two environmental monitoring points; a normalization processing unit, configured to normalize the calculation result to obtain a correlation strength value between any two environmental monitoring points and form a correlation strength value matrix.

10. The mobile communication based environment monitoring alarm platform as claimed in claim 8, wherein, The real-time derived data construction module specifically includes: a prediction value analysis unit, configured to analyze a monitoring prediction value of each associated environmental monitoring point for the target environmental monitoring point based on the monitoring data of each associated environmental monitoring point of the target environmental monitoring point; a comprehensive fitting unit, configured to comprehensively fit the monitoring prediction values of all associated environmental monitoring points for the target environmental monitoring point based on the correlation strength values of each associated environmental monitoring point in the associated neighbor set to obtain the real-time derived data of the environmental verification monitoring point corresponding to the environmental monitoring point.

Images