A monitoring and early warning method for rapid and accurate source tracing

Abstract

This invention provides a monitoring and early warning method for rapid and accurate source tracing, relating to the field of deep source tracing technology for environmental pollution problems. The method includes: acquiring enterprise status monitoring data through edge layer monitoring equipment and acquiring regional environmental status monitoring data and enterprise attribute information through cloud devices; obtaining alarm information; screening related enterprises and determining the source tracing range impact value and wind field impact value of related enterprises; obtaining the risk database score of related enterprises; determining the impact value of the related enterprises' operating enterprises; determining the credibility of multi-stage verification within the enterprise; obtaining the comprehensive impact value of operating enterprises; and determining the risk level of related enterprises. According to this invention, a multi-scale, multi-level source tracing system driven by multi-source data fusion and cloud-edge collaboration can be constructed, enabling precise location and rapid response of pollution sources from the region to the enterprise and workshop levels, timely identification of the causes and sources of responsibility for environmental degradation, and providing scientific support for emergency response to atmospheric events and improvement of air quality.

Description

Technical Field

[0001] This invention relates to the field of in-depth source tracing technology for environmental pollution problems, and in particular to a monitoring and early warning method for rapid and accurate source tracing. Background Technology

[0002] Currently, the mainstream technologies for tracing the source of air pollution mainly include: diffusion model methods (e.g., CALPUFF, AERMOD, etc.), which can simulate the transport and diffusion of pollutants based on meteorological data and pollution source emission inventories, and inversely extrapolate potential source areas through Gaussian diffusion or Lagrange particle models. However, this method is highly dependent on the accuracy of input parameters (e.g., source strength, meteorological field, etc.), and has significant errors in complex terrain or real-time changing scenarios. Furthermore, it is computationally time-consuming and cannot meet the needs of minute-level real-time source tracing. Receptor model methods (e.g., positive definite matrix factorization (PMF), chemical mass balance (CMB, etc.)), which analyze the chemical composition spectrum of particulate matter collected from environmental receptor points and match it with the known pollution source composition spectrum (source spectrum library) to calculate the contribution rate of each source. This method relies on extensive offline sampling and laboratory analysis, resulting in long cycles and high costs. Insufficient localization of the source spectrum library severely impacts accuracy, hindering rapid response and workshop-level location. Fingerprint database tracing technology identifies pollution sources by establishing a unique "fingerprint" feature database of pollutant emissions (e.g., VOC component ratios, isotopic characteristics), collecting environmental samples at the time of pollution occurrence, and matching spectral / mass spectrometry features. While this method has been applied in VOCs and soil pollution tracing, it has significant limitations: First, it heavily relies on a pre-established, comprehensive fingerprint database, which is costly and labor-intensive to build and maintain, making it difficult to cover all potential pollution sources and process changes. Second, pollutants in the environment undergo chemical transformation and dilution during transport, leading to deviations between the measured spectrum and the standard fingerprint in the source spectrum library, reducing matching accuracy. Third, it typically requires offline sampling and laboratory analysis, a cumbersome and time-consuming process that cannot meet the demands of real-time, rapid response tracing. Traditional monitoring combined with manual inspection: Commonly used methods rely on limited online monitoring stations triggering alarms, followed by on-site investigations by environmental personnel using portable equipment. This method suffers from slow response speed, high labor costs, and heavy reliance on personal experience, making it difficult to effectively locate concealed and intermittent emission sources in complex industrial areas. In related technologies, model calculations or manual investigations typically take hours or even days, failing to capture short-term, sudden emission events such as instantaneous illegal discharges or abnormal start-up and shutdown of pollution control facilities. Furthermore, it relies excessively on environmental concentration data or limited source inventories, failing to deeply integrate with dynamic data such as enterprise production conditions, real-time operating status of pollution control facilities (e.g., current, voltage, valve opening), and energy consumption. It lacks monitoring of the pollution generation process, and in complex scenarios such as multi-source intersections and sudden changes in meteorological conditions, the contribution rate calculation error of traditional methods can exceed 30%. It typically only traces pollution to the park or enterprise level, unable to accurately pinpoint specific workshops, production lines, or equipment. Therefore, it suffers from problems such as single data dimension, low spatiotemporal resolution, and insufficient source tracing accuracy, making it difficult to deeply integrate multi-source data and achieve real-time accurate source tracing, resulting in low accuracy and efficiency in source tracing.

[0003] The information disclosed in the background section of this application is intended only to enhance the understanding of the general background of this application and should not be construed as an admission or in any way implying that the information constitutes prior art known to those skilled in the art. Summary of the Invention

[0004] This invention provides a monitoring and early warning method for rapid and accurate source tracing, which can solve the technical problems of related technologies being unable to deeply integrate multi-source data and being unable to conduct real-time and accurate source tracing.

[0005] According to a first aspect of the present invention, a monitoring and early warning method for rapid and accurate source tracing is provided, comprising:

[0006] Enterprise status monitoring data is obtained through edge layer monitoring devices, and regional environmental status monitoring data and enterprise attribute information are obtained through cloud devices.

[0007] By analyzing enterprise status monitoring data and regional environmental status monitoring data through edge layer monitoring devices and cloud devices, alarm information can be obtained;

[0008] Based on alarm information and regional environmental monitoring data, related enterprises were screened, and the source tracing impact value and wind field impact value of the related enterprises were determined.

[0009] Based on enterprise status monitoring data and enterprise attribute information, obtain the risk database score of related enterprises;

[0010] Based on the traceability scope impact value and wind field impact value of related enterprises, as well as the risk database score, the impact value of the related enterprises' operating enterprises is determined;

[0011] Based on the enterprise status monitoring data of related enterprises, determine the credibility of multi-stage verification within the enterprise;

[0012] The comprehensive impact value of the enterprise in production is obtained based on the impact value of the enterprise in production and the credibility of multi-stage verification within the enterprise.

[0013] The risk level of related enterprises is determined based on the comprehensive impact value of enterprises in operation.

[0014] According to the present invention, screening related enterprises and determining the traceability range impact value and wind field impact value of related enterprises includes:

[0015] Based on the alarm information, determine the abnormal location information;

[0016] Based on the abnormal location information, screen the potential related enterprises and determine the traceability scope and impact value of the potential related enterprises;

[0017] Based on regional environmental monitoring data, related companies were screened from the list of potential related companies, and the wind field impact values ​​of the related companies were determined.

[0018] According to the present invention, screening potential related enterprises and determining the traceability scope impact value of potential related enterprises includes:

[0019] Set the anomaly filtering range with the anomaly location information as the center;

[0020] Companies within the anomaly screening range will be identified as pending related companies.

[0021] Obtain the distance between the pending related enterprises and the abnormal location information;

[0022] According to the formula

[0023]

[0024] Determine the impact value of the traceability scope of the unidentified related enterprises Where A is the influence coefficient of the traceability range constraint factor, and X is the distance between the undetermined related enterprise and the abnormal location information.

[0025] According to the present invention, screening related enterprises from the undetermined related enterprises and determining the wind field impact value of the related enterprises includes:

[0026] The pollutant transport path was obtained using the Lagrange inverse trajectory model and anomaly location information;

[0027] Set the sector-shaped screening range according to the pollutant transport path;

[0028] The undetermined related companies within the fan-shaped screening range are identified as related companies;

[0029] Obtain the center line of the sector, and the angle between the line connecting the associated enterprise and the abnormal location information and the center line;

[0030] According to the formula

[0031]

[0032] Determine the wind field impact value of related enterprises Where B is the wind field trajectory constraint influence coefficient. The angle between the line connecting the associated enterprise and the abnormal location information and the center line.

[0033] According to the present invention, obtaining the risk database score of related enterprises includes:

[0034] Based on enterprise status monitoring data and the scores of various alarm events, determine the alarm indicator scores of related enterprises;

[0035] Based on the enterprise status monitoring data and the scores of various enterprise investigation statuses, the scores of the investigation hazard indicators for related enterprises are determined;

[0036] Based on enterprise status monitoring data and various enterprise emergency environmental incident risk scores, determine the enterprise emergency environmental incident risk classification index scores of related enterprises;

[0037] The risk database score of related enterprises is determined based on the alarm indicator score, the hidden danger investigation indicator score, the enterprise sudden environmental incident risk classification indicator score, and the preset first weight.

[0038] According to the present invention, determining the risk classification index score for sudden environmental incidents of related enterprises includes:

[0039] Based on the risk material data, production processes and atmospheric environmental risk control levels of related enterprises, and the sensitivity of atmospheric environmental risk receptors, risk scores for various types of sudden environmental events are determined.

[0040] According to the present invention, the impact value of the related enterprises' production enterprises is determined based on the traceability scope impact value and wind field impact value of the related enterprises, as well as the risk database score, including:

[0041] Based on the enterprise status monitoring data of related enterprises, determine the alarm impact value of related enterprises;

[0042] Based on the alarm impact value, the source tracing range impact value, the wind field impact value, and the risk database score, the impact value of the related enterprises' operating enterprises is determined.

[0043] According to the present invention, determining the reliability of multi-stage verification within an enterprise includes:

[0044] Based on the enterprise status monitoring data of related enterprises, determine the highest frequency coefficient of anomalies in each link of the related enterprises;

[0045] According to the formula

[0046]

[0047] Determine the verification coefficient for the i-th step of the j-th related enterprise. ,in, Let be the intensity coefficient of the anomaly occurring in the i-th stage of the j-th enterprise. Let be the weight coefficient of the i-th stage of the j-th enterprise. Let be the highest frequency coefficient of anomalies in the ith stage of the j-th enterprise;

[0048] According to the formula

[0049]

[0050] Obtain the multi-stage comprehensive verification coefficient of the j-th related enterprise. Where G is the cross-environment consistency coefficient. The number of links for the j-th related enterprise;

[0051] According to the formula

[0052]

[0053] Determine the reliability of multi-stage verification within the enterprise of the j-th related enterprise. ,in, This is the comprehensive verification coefficient for the first related enterprise across multiple stages. This is the comprehensive verification coefficient for the second related enterprise across multiple stages. This is the comprehensive verification coefficient for the nth related enterprise across multiple stages.

[0054] According to the present invention, obtaining the comprehensive impact value of an operating enterprise includes:

[0055] According to the formula

[0056]

[0057] Obtain the comprehensive impact value EE of the related enterprise's operating enterprises, where DE is the impact value of the related enterprise's operating enterprises, and S is the credibility of the multi-stage verification within the related enterprise.

[0058] According to a second aspect of the present invention, a monitoring and early warning system for rapid and accurate source tracing is provided, comprising:

[0059] The acquisition module acquires enterprise status monitoring data through edge layer monitoring devices and regional environmental status monitoring data and enterprise attribute information through cloud devices.

[0060] The alarm information module analyzes enterprise status monitoring data and regional environmental status monitoring data through edge layer monitoring devices and cloud devices to obtain alarm information;

[0061] The filtering and impact value module filters related enterprises based on alarm information and regional environmental monitoring data, and determines the source range impact value and wind field impact value of related enterprises;

[0062] The risk database score module obtains the risk database score of related enterprises based on enterprise status monitoring data and enterprise attribute information;

[0063] The "In-Production Enterprise Impact Value" module determines the impact value of the in-production enterprises of related enterprises based on the traceability scope impact value and wind field impact value of related enterprises, as well as the risk database score.

[0064] The credibility module determines the credibility of multiple verification processes within an enterprise based on enterprise status monitoring data from related companies.

[0065] The comprehensive impact value module obtains the comprehensive impact value of the enterprise in production based on the impact value of the enterprise in production and the credibility of multi-stage verification within the enterprise.

[0066] The risk level module determines the risk level of related enterprises based on the comprehensive impact value of enterprises in operation.

[0067] By adopting the above technical solution, the present invention can achieve the following technical effects:

[0068] According to this invention, enterprise status monitoring data can be acquired through edge layer monitoring devices, and regional environmental status monitoring data and enterprise attribute information can be acquired through cloud devices. Alarm information can be obtained through analysis, thereby screening related enterprises and determining the source tracing scope and wind field impact values ​​of these related enterprises. Based on the enterprise status monitoring data and enterprise attribute information, risk database scores of related enterprises are obtained, thereby determining the impact value of the related enterprises' production enterprises. Furthermore, based on the enterprise status monitoring data of related enterprises, the credibility of multi-stage verification within the enterprise is determined, thus obtaining the comprehensive impact value of the production enterprises, and ultimately determining the risk level of the related enterprises. A multi-scale, multi-level source tracing system driven by multi-source data fusion and cloud-edge collaboration can be constructed, enabling rapid, accurate, and automated source tracing of environmental pollution problems outside production enterprises. This achieves precise location and rapid response of pollution sources from the region to the enterprise and workshop levels, timely identification of the causes and sources of responsibility for environmental degradation, and provides scientific support for emergency response to atmospheric events and improvement of air quality. It can acquire enterprise status monitoring data through edge layer monitoring devices and regional environmental status monitoring data and enterprise attribute information through cloud devices, and obtain alarm information after analysis. This breaks through the limitation of traditional methods that can only trace back to the enterprise, and can comprehensively consider data such as the enterprise's internal DCS and electricity consumption for tracing. Furthermore, the cloud-edge collaborative architecture can reasonably allocate computing load, with the edge responsible for real-time early warning and the cloud responsible for in-depth analysis, improving the real-time performance of massive data processing. When determining the comprehensive impact value of enterprises in production, the comprehensive impact value of enterprises in production can be obtained based on the impact value of enterprises in production and the credibility of multiple links within the enterprise. A decision model based on dynamic weight and credibility assessment is designed to automatically evaluate the credibility of different data sources and dynamically adjust their weights in the tracing decision, improving the system's anti-interference ability and the reliability of the results. In addition, a two-level linkage tracing mechanism of periphery and periphery is created. First, the scope is macroscopically delineated through a physical model, and then the microscopic precise positioning is achieved through a data model, so that the system can maintain stable tracing performance and robustness under complex interference. Furthermore, a multi-source data cross-validation and physical mechanism-data-driven fusion model can be created. Through feature engineering and machine learning algorithms, deep correlations between environmental data, operational data, energy data, and video data can be mined. This deep fusion of multi-source data and intelligent correlation analysis improves the efficiency and accuracy of source tracing, enabling rapid identification of pollution causes to guide regulatory authorities in immediately implementing risk control and source remediation. This can significantly enhance the ability to prevent and respond quickly to pollution incidents, reduce the continuous accumulation and diffusion of pollutants in the atmosphere, thereby improving regional air quality and reducing public health risks.

[0069] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only, and are not intended to limit the invention. Other features and aspects of the invention will become clearer from the following detailed description of exemplary embodiments with reference to the accompanying drawings. Attached Figure Description

[0070] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other embodiments can be obtained based on these drawings without creative effort.

[0071] Figure 1 An exemplary flowchart of a monitoring and early warning method for rapid and accurate source tracing according to an embodiment of the present invention is shown.

[0072] Figure 2 An exemplary flowchart is shown below, illustrating the process of screening related enterprises and determining the traceability range impact value and wind field impact value of related enterprises according to an embodiment of the present invention.

[0073] Figure 3 A block diagram of a monitoring and early warning system for rapid and accurate source tracing according to an embodiment of the present invention is shown as an example. Detailed Implementation

[0074] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0075] The technical solution of the present invention will be described in detail below with reference to specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments.

[0076] Figure 1 An exemplary flowchart illustrates a monitoring and early warning method for rapid and accurate source tracing according to an embodiment of the present invention, the method comprising:

[0077] Step S1: Obtain enterprise status monitoring data through edge layer monitoring devices, and obtain regional environmental status monitoring data and enterprise attribute information through cloud devices;

[0078] Step S2: Analyze enterprise status monitoring data and regional environmental status monitoring data through edge layer monitoring devices and cloud devices to obtain alarm information;

[0079] Step S3: Based on the alarm information and regional environmental monitoring data, screen related enterprises and determine the source tracing range impact value and wind field impact value of related enterprises;

[0080] Step S4: Obtain the risk database score of related enterprises based on enterprise status monitoring data and enterprise attribute information;

[0081] Step S5: Determine the impact value of the related enterprises' production enterprises based on the traceability scope impact value and wind field impact value of the related enterprises, as well as the risk database score.

[0082] Step S6: Determine the credibility of multi-stage verification within the enterprise based on the enterprise status monitoring data of the related enterprises;

[0083] Step S7: Based on the impact value of the enterprise in production and the credibility of multi-stage verification within the enterprise, obtain the comprehensive impact value of the enterprise in production;

[0084] Step S8: Determine the risk level of related enterprises based on the comprehensive impact value of enterprises in production.

[0085] According to embodiments of the present invention, a monitoring and early warning method for rapid and accurate source tracing can acquire enterprise status monitoring data through edge layer monitoring devices and regional environmental status monitoring data and enterprise attribute information through cloud devices. This data is then analyzed to obtain alarm information, thereby screening related enterprises and determining the source tracing scope and wind field impact values ​​of these enterprises. Based on the enterprise status monitoring data and enterprise attribute information, risk database scores for related enterprises are obtained, thereby determining the impact value of the related enterprises' production enterprises. Furthermore, based on the enterprise status monitoring data of related enterprises, the credibility of multi-stage verification within the enterprise is determined, resulting in a comprehensive impact value for the production enterprises, thus determining the risk level of the related enterprises. This method can construct a multi-scale, multi-level source tracing system driven by multi-source data fusion and cloud-edge collaboration, enabling rapid, accurate, and automated source tracing of environmental pollution problems outside production enterprises. It achieves precise location and rapid response of pollution sources from the region to the enterprise and workshop levels, promptly identifying the causes and sources of responsibility for environmental degradation, and providing scientific support for emergency response to atmospheric events and the improvement of air quality.

[0086] Example 1:

[0087] According to an embodiment of the present invention, in step S1, enterprise status monitoring data is acquired through edge layer monitoring devices, and regional environmental status monitoring data and enterprise attribute information are acquired through cloud devices. The enterprise status monitoring data includes enterprise production conditions (e.g., start / stop status of key production equipment, production load, material input) and treatment facility parameters (e.g., fan frequency, air volume, reagent dosage, temperature, pressure, etc.), production conditions and treatment facility electricity consumption (e.g., production line electricity meters, dedicated treatment facility electricity meters, etc.), real-time online monitoring data of pollution source emissions (e.g., concentrations and flow rates of VOCs, SO2, NOx, etc. at organized exhaust gas emission outlets), online monitoring data of rainwater discharge outlets, video surveillance, access control data, etc., which are accessed through edge layer monitoring devices (e.g., environmental quality monitoring devices, pollution control facility operation monitoring devices, etc.). Data accessed directly through cloud devices, including enterprise boundary data (e.g., online monitoring data of VOCs / odor at enterprise boundaries), online monitoring data of public areas within the industrial park (i.e., data from public environmental air quality monitoring stations within the park), online monitoring data of sensitive points (e.g., air quality and complaint records of sensitive points around the park), enterprise environmental impact assessments, enterprise risk assessments, enterprise spatial distribution (i.e., enterprise spatial coordinates), meteorological data (e.g., wind speed, wind direction, boundary layer height), mobile monitoring data (e.g., mobile vehicle monitoring data such as TVOC concentration and path trajectory), and data collected through manual investigation, constitutes the environmental status monitoring data and enterprise attribute information for the aforementioned area. Furthermore, these various data can be preprocessed, such as through format conversion, unit standardization, quality verification, and outlier removal, to standardize data from different sources and ensure a unified scale.

[0088] Example 2:

[0089] According to an embodiment of the present invention, in step S2, enterprise status monitoring data and regional environmental status monitoring data are analyzed by edge layer monitoring devices and cloud devices to obtain alarm information. The edge layer monitoring devices can store the collected enterprise status monitoring data and regional environmental status monitoring data at the edge and perform lightweight anomaly analysis (e.g., anomaly analysis of feature vectors of various data). The cloud devices then perform a series of data analyses on key feature quantities (e.g., anomaly markers, statistical values, etc.) that have undergone edge analysis, and transmit the identified suspect objects and suspected links to the closed-loop management module, thereby generating alarm information (e.g., a high TVOC alarm generated by a public ambient air quality monitoring station at 10:00). This forms a collaborative system of rapid edge response and in-depth cloud analysis and closed-loop management. Without reducing the accuracy of the analysis, by analyzing enterprise status monitoring data and regional environmental status monitoring data, the cloud computing power and storage pressure are reduced, improving the overall source tracing response speed and accelerating the overall source tracing and handling efficiency.

[0090] In the example, the edge layer monitoring device's server stores the enterprise's real-time raw data and deploys a lightweight anomaly detection algorithm. This algorithm rapidly screens enterprise status monitoring data, including production conditions, treatment facility operation, and online emission monitoring data, as well as regional environmental status monitoring data. It identifies significant anomalies such as sudden concentration changes, asynchronous production and treatment processes, and abnormal power consumption, generating alarm information and local alarm records. The cloud-based platform only receives key feature quantities reported from the edge layer and maintains the enterprise's historical risk database, long-term emission feature database, and meteorological database for in-depth source tracing analysis.

[0091] This approach allows for the acquisition of enterprise status monitoring data through edge-layer monitoring devices and the acquisition of regional environmental status monitoring data and enterprise attribute information through cloud devices. Analysis of these data leads to alarm information, overcoming the limitations of traditional methods that can only trace back to the enterprise level. It enables comprehensive tracing by considering data from the enterprise's internal DCS and electricity consumption. Furthermore, the cloud-edge collaborative architecture rationally allocates computing load, with the edge handling real-time alerts and the cloud handling in-depth analysis, improving the real-time performance of massive data processing.

[0092] Example 3:

[0093] Figure 2 An exemplary flowchart is shown for screening related enterprises and determining the traceability range impact value and wind field impact value of related enterprises according to an embodiment of the present invention.

[0094] According to an embodiment of the present invention, in step S3, related enterprises are screened based on alarm information and regional environmental monitoring data, and the traceability range impact value and wind field impact value of the related enterprises are determined, including: step S31, determining abnormal location information based on alarm information; step S32, screening pending related enterprises based on abnormal location information, and determining the traceability range impact value of the pending related enterprises; step S33, screening related enterprises from the pending related enterprises based on regional environmental monitoring data, and determining the wind field impact value of the related enterprises.

[0095] According to an embodiment of the present invention, in step S31, abnormal location information is determined based on the alarm information. The location (e.g., coordinates) of the problem points of external environmental pollution of the production enterprise, such as high-value alarms from public area ambient air quality monitoring stations, odor complaints from sensitive points, and abnormally high values ​​from mobile monitoring on park roads, can be used as abnormal location information. For example, if the alarm information is that a public ambient air quality monitoring station generates a high-value TVOC alarm at 10:00, the location of the public ambient air quality monitoring station is the abnormal location information.

[0096] Example 4:

[0097] According to an embodiment of the present invention, in step S32, based on the anomaly location information, screening pending related enterprises and determining the traceability range impact value of the pending related enterprises includes: setting an anomaly screening range with the anomaly location information as the center; identifying enterprises within the anomaly screening range as pending related enterprises; obtaining the distance between the pending related enterprises and the anomaly location information; and determining the traceability range impact value of the pending related enterprises according to formula (1). ,

[0098] (1)

[0099] Where A is the influence coefficient of the traceability range constraint factor, and X is the distance between the undetermined related enterprise and the abnormal location information.

[0100] According to an embodiment of the present invention, an anomaly screening range is set with the anomaly location information as the center. For example, a circular range with a radius of 5 km centered on the anomaly location information is set as the anomaly screening range. Further, enterprises within the anomaly screening range can be identified as potential related enterprises. Enterprises outside the anomaly screening range are directly excluded from this tracing scope, thereby eliminating low-probability objects with no spatial connection to the anomaly.

[0101] According to an embodiment of the present invention, in formula (1), A is the influence coefficient of the traceability range constraint factor, for example, A is 100. Further, the ratio of the influence coefficient of the traceability range constraint factor to the distance between the pending related enterprise and the anomaly location information can be used as the traceability range influence value of the pending related enterprise. For example, if the distance between a potential related enterprise and an anomaly location is 1.2 km, and the influence coefficient of the traceability range constraint factor is 100, then the traceability range influence value of this potential related enterprise is 100 / 1.2 = 83.3.

[0102] Example 5:

[0103] According to an embodiment of the present invention, in step S33, based on regional environmental monitoring data, related enterprises are screened from the undetermined related enterprises, and the wind field impact value of the related enterprises is determined, including: obtaining pollutant transport paths through the Lagrange inverse trajectory model and anomaly point information; setting a sector-shaped screening range based on the pollutant transport paths; determining the undetermined related enterprises within the sector-shaped screening range as related enterprises; obtaining the centerline of the sector and the angle between the line connecting the related enterprises and the anomaly point information and the centerline; and determining the wind field impact value of the related enterprises according to formula (2). ,

[0104] (2)

[0105] Where B is the wind field trajectory constraint influence coefficient. The angle between the line connecting the associated enterprise and the abnormal location information and the center line.

[0106] According to an embodiment of the present invention, based on meteorological data such as real-time wind speed, wind direction, and boundary layer height, the transmission path of pollutants before the appearance of anomaly information is simulated using a Lagrange inverse trajectory model, which is the pollutant transmission path. Using the angle bisector of the pollutant transmission path as the center line of the sector, a sector range of ±45° (i.e., a sector range of 90°) is determined, which is the sector screening range. Further, unidentified related enterprises within the sector screening range are identified as related enterprises. Under wind field constraints, unidentified related enterprises exceeding the sector screening range can be eliminated, thereby further narrowing down the set of highly suspicious enterprises. The angle bisector of the pollutant transmission path is the center line of the sector, which can then determine the angle between the line connecting the related enterprise and the anomaly information and the center line.

[0107] According to an embodiment of the present invention, in formula (2), B is the wind field trajectory constraint influence coefficient, for example, B is 100. Further, the wind field trajectory constraint influence coefficient can be used as the ratio of the angle between the line connecting the associated enterprise and the abnormal location information and the centerline, as the wind field influence value of the associated enterprise. For example, if the angle between the line connecting a related enterprise and the abnormal location information and the center line is 10°, and the wind field trajectory constraint influence coefficient is 100, then the wind field influence value of the related enterprise is 100 / 10=10.

[0108] Example 6:

[0109] According to an embodiment of the present invention, in step S4, the risk database score of the associated enterprise is obtained based on enterprise status monitoring data and enterprise attribute information, including: determining the alarm indicator score of the associated enterprise based on the enterprise status monitoring data and the scores of various alarm events; determining the hidden danger indicator score of the associated enterprise based on the enterprise status monitoring data and the scores of various enterprise investigation statuses; determining the enterprise sudden environmental event risk classification indicator score of the associated enterprise based on the enterprise status monitoring data and the risk scores of various enterprise sudden environmental events; and determining the risk database score of the associated enterprise based on the alarm indicator score, the hidden danger indicator score, the enterprise sudden environmental event risk classification indicator score, and a preset first weight.

[0110] According to an embodiment of the present invention, the alarm index score of related enterprises is determined based on enterprise status monitoring data and the scores of various alarm events. The number of various alarm events for each related enterprise is determined based on the enterprise status monitoring data, such as the number of alarms related to falsification of automatic monitoring data over the years, the number of alarms related to abnormal pollution source data, the number of alarms related to abnormal treatment facilities, the number of alarms related to abnormal electricity consumption, and the number of alarms related to abnormal water-air balance. Furthermore, a preset alarm index score can be manually set for each type of alarm event; for example, the preset alarm index score for each alarm event is 2. Further, the product of the total number of various alarm events for the related enterprise and the aforementioned preset alarm index score can be used as the alarm index score for that related enterprise. For example, if the preset alarm index score for each alarm event is 2, and a certain related enterprise has 5 alarms related to falsification of automatic monitoring data over the years, 3 alarms related to abnormal pollution source data, and no other alarm events, then the alarm index score for that related enterprise is (3+5)×2=16.

[0111] According to an embodiment of the present invention, the hazard indicator score for related enterprises is determined based on enterprise status monitoring data and scores of various enterprise inspection statuses. The frequency of various enterprise inspection statuses for each related enterprise can be determined based on the enterprise status monitoring data, such as the number of problems discovered during historical internal hazard inspections, the number of problems discovered during temporary monitoring at the factory boundary, etc. Furthermore, a preset hazard indicator score can be manually set for each type of enterprise inspection status; for example, the preset hazard indicator score for each enterprise inspection status is 3 per inspection. Further, similar to determining the alarm indicator score for related enterprises, the product of the total number of inspections of various enterprise inspection statuses for related enterprises and the aforementioned preset hazard indicator score can be used as the hazard indicator score for that related enterprise.

[0112] Example 7:

[0113] According to an embodiment of the present invention, the risk rating index of the enterprise's sudden environmental incident is determined based on enterprise status monitoring data and various enterprise sudden environmental incident risk scores, including: determining the risk scores of various enterprises' sudden environmental incidents based on the risk material data, production process and atmospheric environmental risk control level, and atmospheric environmental risk receptor sensitivity of the related enterprises.

[0114] According to embodiments of the present invention, the weights of three types of corporate environmental emergencies—risk substance data, production process and atmospheric environmental risk control level, and atmospheric environmental risk receptor sensitivity—can be artificially set. For example, the weight of each of the above three types of corporate environmental emergencies can be set to 1. Furthermore, the risk substance data (i.e., the quantity of air-related risk substances), production process and atmospheric environmental risk control level, and atmospheric environmental risk receptor sensitivity of each related enterprise can be determined based on industry status monitoring data. Further, the corporate environmental emergency risk score can be determined based on the risk substance data, production process and atmospheric environmental risk control level, and atmospheric environmental risk receptor sensitivity. When determining the risk score for a company's emergency environmental incident based on risky substance data, it can be determined according to the ratio of the risky substance data to the critical quantity. For example, when the ratio of the risky substance data to the critical quantity is <1, the corresponding risk score for the company's emergency environmental incident is 0; when 1 ≤ the ratio of the risky substance data to the critical quantity is <10, the corresponding risk score for the company's emergency environmental incident is 5; when 10 ≤ the ratio of the risky substance data to the critical quantity is <50, the corresponding risk score for the company's emergency environmental incident is 15; when 50 ≤ the ratio of the risky substance data to the critical quantity is <100, the corresponding risk score for the company's emergency environmental incident is 25; and when the ratio of the risky substance data to the critical quantity is ≥100, the corresponding risk score for the company's emergency environmental incident is 35.

[0115] According to embodiments of the present invention, when determining the enterprise's emergency environmental incident risk score in relation to the production process and atmospheric environmental risk control level, the score can be determined based on the risky processes and equipment in the related enterprise's production process, atmospheric environmental risk prevention and control measures, and the occurrence of emergency atmospheric environmental incidents. For example, the production process and atmospheric environmental risk control level value obtained with reference to the "Enterprise Emergency Environmental Incident Risk Classification Method" (HJ941-2018) can be used as the enterprise's emergency environmental incident risk score in relation to the production process and atmospheric environmental risk control level. When determining the enterprise's emergency environmental incident risk score based on the atmospheric environmental risk receptor sensitivity level, the score can be determined based on the atmospheric environmental risk receptor sensitivity type of the related enterprise. For example, the enterprise emergency environmental incident risk score for type 1 atmospheric environmental risk receptor sensitivity is 30; the enterprise emergency environmental incident risk score for type 2 atmospheric environmental risk receptor sensitivity is 15; and the enterprise emergency environmental incident risk score for type 3 atmospheric environmental risk receptor sensitivity is 0.

[0116] According to an embodiment of the present invention, based on the weights of the enterprise's sudden environmental events, the enterprise sudden environmental event risk scores of the aforementioned related enterprises—including risk material data, production process and atmospheric environmental risk control level, and atmospheric environmental risk receptor sensitivity—are weighted and summed to obtain the enterprise sudden environmental event risk classification index score. For example, if the weights of the three types of enterprise sudden environmental events are all 1, and the enterprise sudden environmental event risk scores of a certain related enterprise for risk material data, production process and atmospheric environmental risk control level, and atmospheric environmental risk receptor sensitivity are 25, 15, and 15 respectively, then the enterprise sudden environmental event risk classification index score of that related enterprise is 25×1 + 15×1 + 15×1 = 55.

[0117] According to an embodiment of the present invention, the risk database score of an associated enterprise is determined based on the alarm index score, the hidden danger investigation index score, the enterprise sudden environmental event risk classification index score, and a preset first weight. The preset first weights corresponding to the alarm index score, the hidden danger investigation index score, and the enterprise sudden environmental event risk classification index score can be manually set. The risk database score is obtained by weighted summing of these preset first weights. For example, if the preset first weight corresponding to the alarm index score is 0.5, the preset first weight corresponding to the hidden danger investigation index score is 0.2, and the preset first weight corresponding to the enterprise sudden environmental event risk classification index score is 0.1, and the alarm index score, hidden danger investigation index score, and enterprise sudden environmental event risk classification index score of an associated enterprise are 12, 9, and 55 respectively, then the risk database score of that associated enterprise is 12×0.5+9×0.2+55×0.1=13.3.

[0118] Example 8:

[0119] According to an embodiment of the present invention, in step S5, the impact value of the related enterprise's operating enterprises is determined based on the traceability scope impact value and wind field impact value of the related enterprise, as well as the risk database score. This includes: determining the alarm impact value of the related enterprise based on the enterprise status monitoring data of the related enterprise; and determining the impact value of the related enterprise's operating enterprises based on the alarm impact value, traceability scope impact value, wind field impact value, and risk database score.

[0120] According to an embodiment of the present invention, based on the enterprise status monitoring data of related enterprises, the alarm status of production, treatment, and emission data of related enterprises is determined (e.g., the time of abnormalities such as online monitoring alarms, operating condition alarms, pollution control facility alarms, abnormal power consumption, and abnormal water-air balance of related enterprises). Further, based on the time of the abnormality occurrence, a time window of duration T (e.g., 2 hours) is set, and the alarm status of production, treatment, and emission data of related enterprises within T hours before the abnormal time point is retrieved. If an abnormal alarm exists, an alarm impact value is assigned to the related enterprise (e.g., an alarm impact value of 20). For example, if the abnormality occurs at 10:00, the time window T is 2 hours, and a related enterprise experiences alarms such as abnormal temperature of treatment facilities and short-term increase in emission concentration during the period from 09:30 to 09:50, which falls within the period from 08:00 to 10:00, then an alarm impact value of 20 can be assigned to the related enterprise.

[0121] According to an embodiment of the present invention, the impact value of the related enterprise's operating enterprises is determined based on the alarm impact value, the source tracing range impact value, the wind field impact value, and the risk database score. The weights corresponding to the alarm impact value, the source tracing range impact value, the wind field impact value, and the risk database score can be preset manually, and a weighted summation of these values ​​can be performed to obtain the impact value of the related enterprise's operating enterprises. For example, if the weights corresponding to the alarm impact value, the source tracing range impact value, the wind field impact value, and the risk database score are 0.2, 0.2, 0.2, and 0.4, respectively, and the alarm impact value, the source tracing range impact value, the wind field impact value, and the risk database score of a certain related enterprise are 20, 80, 10, and 15, respectively, then the impact value of the related enterprise's operating enterprises is 20×0.2+80×0.2+10×0.2+15×0.4=28. The higher the impact value of the producing enterprise, the higher the correlation between the related enterprise and this pollution event in four dimensions: spatial location, wind direction, historical risk, and abnormal period behavior.

[0122] Example 9:

[0123] According to an embodiment of the present invention, in step S6, the reliability of multi-stage verification within the enterprise is determined based on the enterprise status monitoring data of the related enterprises, including: determining the highest frequency coefficient of anomalies in each stage of the related enterprises based on the enterprise status monitoring data of the related enterprises.

[0124] The verification coefficient of the i-th link of the j-th related enterprise is determined according to formula (3). ,

[0125] (3)

[0126] in, Let be the intensity coefficient of the anomaly occurring in the i-th stage of the j-th enterprise. Let be the weight coefficient of the i-th stage of the j-th enterprise. Let be the highest frequency coefficient of anomalies in the ith stage of the j-th enterprise;

[0127] The multi-stage comprehensive verification coefficient of the j-th related enterprise is obtained according to formula (4). ,

[0128] (4)

[0129] Where G is the cross-environment consistency coefficient. The number of links for the j-th related enterprise;

[0130] The reliability of the multi-stage verification within the enterprise of the j-th related enterprise is determined according to formula (5). ,

[0131] (5)

[0132] in, This is the comprehensive verification coefficient for the first related enterprise across multiple stages. This is the comprehensive verification coefficient for the second related enterprise across multiple stages. This is the comprehensive verification coefficient for the nth related enterprise across multiple stages.

[0133] According to an embodiment of the present invention, the highest frequency coefficient of anomalies in each stage of the related enterprises is determined based on the enterprise status monitoring data of the related enterprises. Based on the enterprise status monitoring data of the related enterprises, the frequency of historical accidents occurring in each stage of the related enterprises is determined, and different highest frequency coefficients of anomalies can be assigned to different frequencies. For example, if the pollution control stage of a related enterprise has a high historical accident frequency, the highest frequency coefficient of anomalies in the pollution control stage can be determined as 0.75; if the pollution treatment stage has a relatively high historical accident frequency, the highest frequency coefficient of anomalies in the pollution treatment stage can be determined as 0.5; if the emission stage has a relatively low historical accident frequency, the highest frequency coefficient of anomalies in the emission stage can be determined as 0.25; and for stages where no accidents have occurred, the highest frequency coefficient of anomalies in the stage can be determined as 0.

[0134] According to an embodiment of the present invention, in formula (3), The intensity coefficient for an anomaly occurring in the i-th stage of the j-th enterprise can be set manually. For example, if the temperature in the treatment stage is abnormal for a short time, the intensity coefficient for the corresponding enterprise's treatment stage can be set to 0.5; if the emission stage rises for a short time but does not exceed the standard, the intensity coefficient for the corresponding enterprise's emission stage can be set to 0.5; if the emission stage slightly exceeds the standard, the intensity coefficient for the corresponding enterprise's emission stage can be set to 0.75; if the pollution control stages are not started and stopped simultaneously, the intensity coefficient for the corresponding enterprise's pollution control stage can be set to 0.75, and so on. The weighting coefficient for the i-th stage of the j-th enterprise can be set manually. For example, the weighting coefficients for the pollution control stage, emission stage, and treatment stage can be set to 1, 1.5, and 1, respectively. Furthermore, the verification coefficient for the i-th stage of the j-th related enterprise can be determined according to formula (3). For example, the intensity coefficient of the abnormality in the emission process of the j-th related enterprise is 0.5, the weighting coefficient is 1.5, and the highest frequency coefficient of the abnormality is 0.75. The verification coefficient of the emission process of the j-th related enterprise... That is, 0.5 × 1.5 × 0.75 = 0.5625. Based on the same processing method, the verification coefficients for each link of the j-th related enterprise can be obtained.

[0135] According to an embodiment of the present invention, in formula (4), G is the cross-environmental consistency coefficient. G is 1 when the relationship between the two links should be positively correlated; otherwise, G is 0. For example, if pollution production increases, pollution control load increases, and emissions decrease (i.e., the relationship is reversed), then the links are aligned, meaning the cross-environmental consistency coefficient G is 1. Further, the sum of the products of the verification coefficients of each link of the j-th related enterprise and the cross-environmental consistency coefficient can be obtained. That is, the multi-stage comprehensive verification coefficient of the j-th related enterprise. .

[0136] According to an embodiment of the present invention, in formula (5), This represents the sum of the multi-stage comprehensive verification coefficients of all related enterprises. Therefore, the ratio of the multi-stage comprehensive verification coefficient of the j-th related enterprise to the sum of the multi-stage comprehensive verification coefficients of all the aforementioned related enterprises can be obtained. This refers to the credibility of the multi-stage verification within the enterprise of the j-th related enterprise. The higher the credibility of multi-stage verification within the enterprise, the higher the internal credibility of the j-th related enterprise as the responsible source of this pollution incident. Based on the same processing method, the credibility of multi-stage verification within each related enterprise can be obtained.

[0137] Example 10:

[0138] According to an embodiment of the present invention, in step S7, the comprehensive impact value of the producing enterprise is obtained based on the impact value of the producing enterprise and the reliability of multi-stage verification within the enterprise, including: obtaining the comprehensive impact value EE of the producing enterprise of the related enterprise according to formula (6).

[0139] (6)

[0140] Where DE represents the impact value of the related enterprise's operating enterprises, and S represents the credibility of the multi-stage verification within the related enterprise.

[0141] According to an embodiment of the present invention, the sum of the credibility of multi-stage verification within the enterprise of 1 and related enterprises can be considered as an amplification coefficient of the influence value of the production enterprises of related enterprises. When S is high, the influence value of the production enterprises of related enterprises can be amplified, thereby obtaining a higher comprehensive influence value of the production enterprises of related enterprises, highlighting related enterprises with matching external conditions and sufficient internal evidence. Conversely, when S is low, even if the influence value of the production enterprises of related enterprises is high, the obtained comprehensive influence value of the production enterprises of related enterprises will not be too high, so as to reduce the probability of misjudgment due to accidental spatial overlap. The influence value of the production enterprises of related enterprises (i.e., external conditions) and the credibility of multi-stage verification within the enterprise (i.e., internal evidence) can be comprehensively considered to determine the comprehensive influence value of the production enterprises of related enterprises, thereby improving the accuracy and comprehensiveness of the determination of the comprehensive influence value of the production enterprises of related enterprises, and used to pinpoint the source of responsibility for pollution incidents.

[0142] In this way, the comprehensive impact value of enterprises in production can be obtained based on the impact value of enterprises in production and the credibility of verification of multiple links within the enterprise. A decision model based on dynamic weight and credibility assessment is designed to automatically evaluate the credibility of different data sources and dynamically adjust their weights in the traceability decision, thereby improving the system's anti-interference ability and the reliability of the results. Furthermore, a two-level linkage traceability mechanism of periphery and periphery is created. First, the scope is macroscopically defined through the physical model, and then the microscopic precise positioning is achieved through the data model, so that the system can still maintain stable traceability performance and robustness under complex interference.

[0143] Example 11:

[0144] According to an embodiment of the present invention, in step S8, the risk level of related enterprises is determined based on the comprehensive impact value of the enterprises in production. The impact values ​​of the enterprises in production of each related enterprise can be ranked from largest to smallest, and divided into three risk levels—major suspicion, significant suspicion, and general suspicion—based on percentage distribution. This is used to determine the related enterprises and suspected links with major suspicion and the highest suspicion. For example, the top 10% of risk levels are considered major suspicion, 10% to 40% are considered significant suspicion, and 40% to 100% are considered general suspicion. In this example, major and significant suspicion targets can be automatically pushed to the closed-loop management module to generate on-site verification tasks and issue key evidence chain reports. This facilitates on-site verification of the traceability results. After regulatory verification, enterprises need to carry out risk rectification based on the verification results and upload the rectification status to the cloud. After regulatory review, closed-loop management of the traceability results is achieved.

[0145] This approach enables the creation of multi-source data cross-validation and physical mechanism-data-driven fusion models. Through feature engineering and machine learning algorithms, it uncovers deep correlations between environmental, operational, energy, and video data. By deeply integrating multi-source data and performing intelligent correlation analysis, it improves the efficiency and accuracy of source tracing, thereby quickly identifying the causes of pollution and guiding regulatory authorities to immediately implement risk control and source remediation. This significantly enhances the ability to prevent and respond quickly to pollution incidents, reduces the continuous accumulation and diffusion of pollutants in the atmosphere, and thus improves regional air quality and reduces public health risks.

[0146] According to embodiments of the present invention, a monitoring and early warning method for rapid and accurate source tracing can acquire enterprise status monitoring data through edge layer monitoring devices and regional environmental status monitoring data and enterprise attribute information through cloud devices. This data is then analyzed to obtain alarm information, thereby screening related enterprises and determining the source tracing scope and wind field impact values ​​of these enterprises. Based on the enterprise status monitoring data and enterprise attribute information, risk database scores for related enterprises are obtained, thereby determining the impact value of the related enterprises' production enterprises. Furthermore, based on the enterprise status monitoring data of related enterprises, the credibility of multi-stage verification within the enterprise is determined, resulting in a comprehensive impact value for the production enterprises, thus determining the risk level of the related enterprises. This method can construct a multi-scale, multi-level source tracing system driven by multi-source data fusion and cloud-edge collaboration, enabling rapid, accurate, and automated source tracing of environmental pollution problems outside production enterprises. It achieves precise location and rapid response of pollution sources from the region to the enterprise and workshop levels, promptly identifying the causes and sources of responsibility for environmental degradation, and providing scientific support for emergency response to atmospheric events and the improvement of air quality. It can acquire enterprise status monitoring data through edge layer monitoring devices and regional environmental status monitoring data and enterprise attribute information through cloud devices. After analysis, alarm information is obtained, breaking through the limitation of traditional methods that can only trace back to the enterprise. It can comprehensively consider data such as the enterprise's internal DCS and electricity consumption for source tracing. Furthermore, the cloud-edge collaborative architecture can reasonably allocate computing load, with the edge responsible for real-time early warning and the cloud responsible for in-depth analysis, improving the real-time performance of massive data processing (reducing the time from pollution event occurrence to output source tracing results from the traditional hours to minutes (less than 3 minutes), achieving near real-time response). When determining the comprehensive impact value of enterprises in production, the comprehensive impact value of enterprises in production can be obtained based on the impact value of enterprises in production and the credibility of verification of multiple links within the enterprise. A decision model based on dynamic weight and credibility assessment is designed to automatically assess the credibility of different data sources and dynamically adjust their weights in the traceability decision, thereby improving the system's anti-interference ability and the reliability of the results. Furthermore, a two-level linkage traceability mechanism of periphery and periphery is created. First, the scope is macroscopically delineated through a physical model, and then the microscopic precise positioning is achieved through a data model, so that the system can still maintain stable traceability performance and robustness under complex interference. Furthermore, a multi-source data cross-validation and physical mechanism-data-driven fusion model can be created. Through feature engineering and machine learning algorithms, deep correlations between environmental data, operating condition data, energy data, and video data can be mined. This deep fusion of multi-source data and intelligent correlation analysis improves the efficiency and accuracy of source tracing (increasing the accuracy of source tracing in complex scenarios from less than 60% in traditional methods to over 90%, with a false alarm rate of <5% and a false negative rate of <1%). This allows for the rapid identification of pollution causes, guiding regulatory authorities to immediately carry out risk control and source remediation. By combining data from the enterprise's internal DCS and electricity consumption, the source can be traced back to specific production workshops, process links, or even individual pieces of equipment, achieving "targeted" precise control.It can significantly improve the ability to prevent and respond quickly to pollution incidents, reduce the continuous accumulation and diffusion of pollutants in the atmosphere, thereby improving regional air quality and reducing public health risks.

[0147] Example 12:

[0148] Figure 3 An exemplary block diagram of a monitoring and early warning system for rapid and accurate source tracing according to an embodiment of the present invention is shown, the system comprising:

[0149] The acquisition module acquires enterprise status monitoring data through edge layer monitoring devices and regional environmental status monitoring data and enterprise attribute information through cloud devices.

[0150] The alarm information module analyzes enterprise status monitoring data and regional environmental status monitoring data through edge layer monitoring devices and cloud devices to obtain alarm information;

[0151] The filtering and impact value module filters related enterprises based on alarm information and regional environmental monitoring data, and determines the source range impact value and wind field impact value of related enterprises;

[0152] The risk database score module obtains the risk database score of related enterprises based on enterprise status monitoring data and enterprise attribute information;

[0153] The "In-Production Enterprise Impact Value" module determines the impact value of the in-production enterprises of related enterprises based on the traceability scope impact value and wind field impact value of related enterprises, as well as the risk database score.

[0154] The credibility module determines the credibility of multiple verification processes within an enterprise based on enterprise status monitoring data from related companies.

[0155] The comprehensive impact value module obtains the comprehensive impact value of the enterprise in production based on the impact value of the enterprise in production and the credibility of multi-stage verification within the enterprise.

[0156] The risk level module determines the risk level of related enterprises based on the comprehensive impact value of enterprises in operation.

[0157] This invention can be a method, apparatus, system, and / or computer program product. The computer program product may include a computer-readable storage medium having computer-readable program instructions loaded thereon for performing various aspects of the invention.

[0158] Those skilled in the art should understand that the embodiments of the present invention described above and shown in the accompanying drawings are merely examples and do not limit the present invention. The objectives of the present invention have been fully and effectively achieved. The functions and structural principles of the present invention have been demonstrated and explained in the embodiments, and any variations or modifications may be made to the implementation of the present invention without departing from the stated principles.

[0159] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A monitoring and early warning method for rapid and accurate source tracing, characterized in that, include: Enterprise status monitoring data is obtained through edge layer monitoring devices, and regional environmental status monitoring data and enterprise attribute information are obtained through cloud devices. By analyzing enterprise status monitoring data and regional environmental status monitoring data through edge layer monitoring devices and cloud devices, alarm information can be obtained; Based on alarm information and regional environmental monitoring data, related enterprises were screened, and the source tracing impact value and wind field impact value of related enterprises were determined. Based on enterprise status monitoring data and enterprise attribute information, obtain the risk database score of related enterprises; Based on the traceability scope impact value and wind field impact value of related enterprises, as well as the risk database score, the impact value of the related enterprises' operating enterprises is determined; Based on the enterprise status monitoring data of related enterprises, determine the credibility of multi-stage verification within the enterprise; The comprehensive impact value of the enterprise in production is obtained based on the impact value of the enterprise in production and the credibility of multi-stage verification within the enterprise. The risk level of related enterprises is determined based on the comprehensive impact value of enterprises in operation; Based on the impact value of the producing enterprise and the reliability of multi-stage verification within the enterprise, the comprehensive impact value of the producing enterprise is obtained, including: According to the formula Obtain the comprehensive impact value EE of the related enterprise's operating enterprises, where DE is the impact value of the related enterprise's operating enterprises, and S is the credibility of the multi-stage verification within the related enterprise.

2. The monitoring and early warning method for rapid and accurate source tracing according to claim 1, characterized in that, Based on alarm information and regional environmental monitoring data, related enterprises were screened, and the source tracing impact value and wind field impact value of these enterprises were determined, including: Based on the alarm information, determine the abnormal location information; Based on the abnormal location information, screen the potential related enterprises and determine the traceability scope and impact value of the potential related enterprises; Based on regional environmental monitoring data, related companies were screened from the list of potential related companies, and the wind field impact values ​​of the related companies were determined.

3. The monitoring and early warning method for rapid and accurate source tracing according to claim 2, characterized in that, Based on the anomaly location information, we screen potential related enterprises and determine the impact value of the traceability scope of these enterprises, including: Set the anomaly filtering range with the anomaly location information as the center; Companies within the anomaly screening range will be identified as pending related companies. Obtain the distance between the pending related enterprises and the abnormal location information; According to the formula Determine the impact value of the traceability scope of the unidentified related enterprises Where A is the influence coefficient of the traceability range constraint factor, and X is the distance between the undetermined related enterprise and the abnormal location information.

4. The monitoring and early warning method for rapid and accurate source tracing according to claim 2, characterized in that, Based on regional environmental monitoring data, related enterprises were screened from the list of potential related enterprises, and the wind field impact values ​​of these related enterprises were determined, including: The pollutant transport path was obtained using the Lagrange inverse trajectory model and anomaly location information; Set the sector-shaped screening range according to the pollutant transport path; The undetermined related companies within the fan-shaped screening range are identified as related companies; Obtain the center line of the sector, and the angle between the line connecting the associated enterprise and the abnormal location information and the center line; According to the formula Determine the wind field impact value of related enterprises Where B is the wind field trajectory constraint influence coefficient. The angle between the line connecting the associated enterprise and the abnormal location information and the center line.

5. The monitoring and early warning method for rapid and accurate source tracing according to claim 1, characterized in that, Based on enterprise status monitoring data and enterprise attribute information, obtain the risk database score of related enterprises, including: Based on enterprise status monitoring data and the scores of various alarm events, determine the alarm indicator scores of related enterprises; Based on the enterprise status monitoring data and the scores of various enterprise investigation statuses, the scores of the investigation hazard indicators for related enterprises are determined; Based on enterprise status monitoring data and various enterprise emergency environmental incident risk scores, determine the enterprise emergency environmental incident risk classification index scores of related enterprises; The risk database score of related enterprises is determined based on the alarm indicator score, the hidden danger investigation indicator score, the enterprise sudden environmental incident risk classification indicator score, and the preset first weight.

6. The monitoring and early warning method for rapid and accurate source tracing according to claim 5, characterized in that, Based on enterprise status monitoring data and various enterprise emergency environmental incident risk scores, the enterprise emergency environmental incident risk classification index scores for related enterprises are determined, including: Based on the risk material data, production processes and atmospheric environmental risk control levels of related enterprises, and the sensitivity of atmospheric environmental risk receptors, risk scores for various types of sudden environmental events are determined.

7. The monitoring and early warning method for rapid and accurate source tracing according to claim 1, characterized in that, Based on the source tracing impact value and wind field impact value of related enterprises, as well as the risk database score, the impact value of the related enterprises' operating enterprises is determined, including: Based on the enterprise status monitoring data of related enterprises, determine the alarm impact value of related enterprises; Based on the alarm impact value, the source tracing range impact value, the wind field impact value, and the risk database score, the impact value of the related enterprises' operating enterprises is determined.

8. The monitoring and early warning method for rapid and accurate source tracing according to claim 1, characterized in that, Based on the enterprise status monitoring data of related enterprises, the reliability of multi-stage verification within the enterprise is determined, including: Based on the enterprise status monitoring data of related enterprises, determine the highest frequency coefficient of anomalies in each link of the related enterprises; According to the formula Determine the verification coefficient for the i-th step of the j-th related enterprise. ,in, Let be the intensity coefficient of the anomaly occurring in the i-th stage of the j-th enterprise. Let be the weight coefficient of the i-th stage of the j-th enterprise. Let be the highest frequency coefficient of anomalies in the i-th stage of the j-th enterprise; According to the formula Obtain the multi-stage comprehensive verification coefficient of the j-th related enterprise. Where G is the cross-environment consistency coefficient. The number of links for the j-th related enterprise; According to the formula Determine the reliability of multi-stage verification within the enterprise of the j-th related enterprise. ,in, This is the comprehensive verification coefficient for the first related enterprise across multiple stages. This is the comprehensive verification coefficient for the second related enterprise across multiple stages. This is the comprehensive verification coefficient for the nth related enterprise across multiple stages.

9. A monitoring and early warning system for rapid and accurate source tracing, characterized in that, include: The acquisition module acquires enterprise status monitoring data through edge layer monitoring devices and regional environmental status monitoring data and enterprise attribute information through cloud devices. The alarm information module analyzes enterprise status monitoring data and regional environmental status monitoring data through edge layer monitoring devices and cloud devices to obtain alarm information; The filtering and impact value module filters related enterprises based on alarm information and regional environmental monitoring data, and determines the source range impact value and wind field impact value of related enterprises; The risk database score module obtains the risk database score of related enterprises based on enterprise status monitoring data and enterprise attribute information; The "In-Production Enterprise Impact Value" module determines the impact value of the in-production enterprises of related enterprises based on the traceability scope impact value and wind field impact value of related enterprises, as well as the risk database score. The credibility module determines the credibility of multiple verification processes within an enterprise based on enterprise status monitoring data from related companies. The comprehensive impact value module obtains the comprehensive impact value of the enterprise in production based on the impact value of the enterprise in production and the credibility of multi-stage verification within the enterprise. The risk rating module determines the risk rating of related enterprises based on the comprehensive impact value of enterprises in operation. Based on the impact value of the producing enterprise and the reliability of multi-stage verification within the enterprise, the comprehensive impact value of the producing enterprise is obtained, including: According to the formula Obtain the comprehensive impact value EE of the related enterprise's operating enterprises, where DE is the impact value of the related enterprise's operating enterprises, and S is the credibility of the multi-stage verification within the related enterprise.

Images