Environmental pollution source intelligent environmental protection monitoring analysis system and method based on Internet of Things

Abstract

The invention relates to the technical field of environmental protection monitoring and analysis systems, in particular to an intelligent environmental protection monitoring and analysis system and method for environmental pollution sources based on the Internet of Things, and the system comprises an industrial park area planning module which is used for collecting and processing enterprise information in an area; the air monitoring and early warning module is used for collecting and analyzing air quality data and performing early warning; the park communication management module is used for realizing data interaction among the components in the system; and the server is used for constructing an environmental pollution source model and carrying out prediction and early warning, and enterprise production process information provided by the industrial park area planning module can be combined with pollutant data detected by the air monitoring and early warning module, so that potential pollution sources can be identified more accurately. The collaborative analysis greatly improves the pollution source identification and early warning capability of the system.

Description

Technical Field

[0001] The present invention relates to the technical field of environmental monitoring and analysis systems, and more specifically, to an intelligent environmental monitoring and analysis system and method for environmental pollution sources based on the Internet of Things. Background Art

[0002] In recent years, with the acceleration of industrialization, environmental pollution has become increasingly serious, especially in industrial parks, where the emission of various pollutants has posed a huge threat to the surrounding environment and public health. Traditional environmental monitoring methods often have problems such as few monitoring points, slow data updates, and limited analysis capabilities, which are difficult to meet the needs of modern environmental management.

[0003] With the development of IoT technology, some IoT-based environmental monitoring systems have begun to emerge. These systems have achieved real-time monitoring of environmental factors such as air quality, water quality, and noise by deploying a large number of sensors in industrial parks. However, most of these systems remain at the level of data collection and simple display, lacking the ability to deeply analyze and intelligently process data. For example, although some systems can display the pollutant concentration of each monitoring point in real time, they cannot effectively identify the pollution source, nor can they predict the pollution trend, let alone provide strong support for environmental management decisions.

[0004] On the other hand, some research institutions have developed complex environmental models that attempt to predict the spread and impact of pollutants through mathematical simulations. Although these models are valuable in theory, they are often too complex and computationally intensive to be widely used in actual environmental management. Moreover, these models usually rely on a large number of assumptions and are difficult to accurately reflect the complexity and variability of actual conditions.

[0005] In addition, the existing environmental monitoring system generally has the problem of "information islands". Environmental data, corporate information, meteorological data and other related information are often scattered in different systems, making it difficult to achieve effective integration and collaborative analysis. This makes it difficult for environmental management departments to fully and accurately grasp the environmental situation and make scientific decisions quickly.

[0006] In the face of these problems, there is an urgent need for an intelligent environmental monitoring and analysis system that can integrate Internet of Things technology, big data analysis and artificial intelligence algorithms. Such a system should be able to achieve comprehensive and real-time monitoring of the industrial park environment, deeply analyze pollution data, accurately identify and locate pollution sources, predict pollution trends, and provide timely and powerful support for environmental management decisions. Summary of the invention

[0007] The present invention is proposed to solve the above technical problems. The present invention provides an intelligent environmental monitoring and analysis system and method for environmental pollution sources based on the Internet of Things, which includes an industrial park regional planning module, an air monitoring and early warning module, a park communication management module and a server. This system architecture not only realizes the comprehensive collection and real-time transmission of environmental data, but also realizes the deep mining and intelligent processing of data through advanced analysis algorithms.

[0008] In order to solve the above technical problems, the present invention adopts the following technical solutions:

[0009] An intelligent environmental monitoring and analysis system for environmental pollution sources based on the Internet of Things, the system includes:

[0010] Industrial park regional planning module, used to collect and process enterprise information in the area;

[0011] Air monitoring and early warning module, used to collect and analyze air quality data and issue early warnings;

[0012] The campus communication management module is used to realize data interaction between components in the system;

[0013] Server, used to build environmental pollution source models and conduct predictions and early warnings.

[0014] Preferably, the industrial park area planning module includes:

[0015] Enterprise type collection unit, used to collect category information of enterprises in the area;

[0016] Enterprise production process collection unit, used to collect enterprise production process;

[0017] The regional planning processing unit is used to analyze the processes that affect regional environmental pollution sources based on enterprise categories and production processes.

[0018] Preferably, the air monitoring and early warning module comprises:

[0019] Air quality collection unit, used to collect air quality parameters in the area in real time;

[0020] An air quality analysis unit for analyzing air quality parameters;

[0021] The air quality early warning unit is used to issue environmental early warnings based on the analysis results.

[0022] Preferably, the air quality analysis unit performs air quality parameter analysis based on the following steps:

[0023] Step A: construct a multidimensional gas concentration matrix;

[0024] Step B: performing gas correlation analysis based on the multidimensional gas concentration matrix;

[0025] Step C: constructing a gas diffusion group theory model based on the gas correlation analysis results;

[0026] Step D: locating the pollution source based on the gas diffusion group theory model;

[0027] Step E: Generate a multi-level warning threshold matrix based on the pollution source positioning result.

[0028] Preferably, the method for constructing the multidimensional gas concentration matrix G in step A comprises:

[0029] Set up an n-dimensional gas sensor array to collect data from m sampling points;

[0030] Organize the collected data into an m×n matrix G:

[0031]

[0032] Among them, g ij is the concentration value of the jth gas at the ith sampling point.

[0033] Preferably, the method for performing gas correlation analysis in step B comprises: calculating a correlation matrix R between gases based on the multidimensional gas concentration matrix G:

[0034]

[0035] Where m is the number of sampling points, 1 is an m-dimensional all-one column vector, G T Represents the transposed matrix of G.

[0036] Preferably, the method for constructing the gas diffusion group theory model in step C comprises:

[0037] Based on the correlation matrix R, a gas diffusion group D is constructed; and the operation of the group D is defined. for

[0038]

[0039] The elements of group D are the eigenvectors of the R matrix, representing the diffusion patterns of different gases.

[0040] Preferably, the method for locating the pollution source in step D comprises: using the elements in the group D to construct a pollution source intensity function S(r, θ):

[0041]

[0042] Among them, J l is the Bessel function of the first kind, α klis its kth zero point, R′ is the radius of the considered area, c kl are unknown coefficients, r and θ represent the radial and angular coordinates in polar coordinates, respectively.

[0043] Preferably, the method for generating a multi-level warning threshold matrix in step E comprises:

[0044] Based on the pollution source intensity function S(r,θ), the warning threshold matrix T is constructed. * :

[0045]

[0046] Among them, q is the number of warning levels, p is the number of pollutant types, and t i j is the threshold value of the i-th level warning for the j-th pollutant.

[0047] An intelligent environmental protection monitoring and analysis method for environmental pollution sources based on the Internet of Things, the method comprising:

[0048] Collect and process enterprise information in the region;

[0049] Collect and analyze air quality data;

[0050] Realize data interaction between components in the system;

[0051] Construct environmental pollution source models and conduct prediction and early warning;

[0052] The analysis process of the air quality data includes the steps described above.

[0053] Compared with the prior art, the beneficial effects of the present invention are embodied in the following aspects:

[0054] First of all, from the perspective of the overall architecture, the system of the present invention realizes intelligent management of the entire process from data collection, transmission, analysis to decision support. The industrial park regional planning module provides the system with detailed enterprise information and production process data, laying the foundation for pollution source identification. The air monitoring and early warning module realizes real-time monitoring and precise analysis of air quality through advanced sensor networks and innovative analysis algorithms. The park communication management module ensures data interaction between various components, breaking the "information island" problem in traditional systems. As the "brain" of the system, the server integrates the data and analysis results of each module and builds a comprehensive environmental pollution source model.

[0055] In terms of module collaboration, the system of the present invention realizes effective fusion and collaborative analysis of multi-source data. For example, the enterprise production process information provided by the industrial park area planning module can be combined with the pollutant data detected by the air monitoring and early warning module to more accurately identify potential pollution sources. This collaborative analysis greatly improves the system's pollution source identification and early warning capabilities.

[0056] In terms of resolving technical contradictions, the system of the present invention cleverly balances the complexity and practicality of analysis. By introducing innovative gas diffusion group theory models and Bessel function expansion methods, the system can accurately describe the complex pollutant diffusion process while maintaining computational efficiency, allowing complex theoretical models to be applied in actual environmental management.

[0057] In terms of complementary and superimposed effects, the system of the present invention achieves refined management of different types of pollutants and different pollution levels through the design of a multi-level warning threshold matrix. This design not only considers the impact of a single pollutant, but also evaluates the combined effects of multiple pollutants, thereby providing a more comprehensive and accurate pollution warning.

[0058] From a microscopic perspective, the invention's innovations in data analysis algorithms, such as multi-dimensional gas concentration matrix construction and gas correlation analysis, have greatly improved the system's data processing capabilities and analysis accuracy. The application of these algorithms enables the system to extract valuable information from massive amounts of monitoring data, providing a scientific basis for environmental management decisions.

[0059] In general, the intelligent environmental monitoring and analysis system for environmental pollution sources based on the Internet of Things provided by the present invention realizes comprehensive, real-time, and intelligent monitoring of the industrial park environment through advanced technical architecture and innovative analysis methods. It can not only timely discover and accurately locate pollution sources, but also predict future pollution trends, providing a powerful decision-making support tool for environmental management departments. The application of this system will significantly improve the environmental management level of industrial parks, effectively reduce environmental pollution risks, and make important contributions to achieving sustainable development. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 It is a flowchart of the overall system of the present invention.

[0061] Figure 2 It is a logic block diagram of the air quality analysis unit of the present invention.

[0062] Figure 3 This is a flow chart of the intelligent environmental monitoring and analysis method for environmental pollution sources based on the Internet of Things of the present invention. DETAILED DESCRIPTION

[0063] like Figure 1-3As shown, the present invention provides an intelligent environmental monitoring and analysis system and method for environmental pollution sources based on the Internet of Things. The system includes an industrial park area planning module 1, an air monitoring and early warning module 2, a park communication management module 3 and a server 4.

[0064] The industrial park regional planning module 1 is used to collect and process enterprise information in the area. This module provides basic data support for subsequent environmental monitoring through comprehensive investigation and analysis of enterprises in the park. Preferably, the industrial park regional planning module 1 includes an enterprise type collection unit 11, an enterprise production process collection unit 12 and a regional planning processing unit 13.

[0065] The enterprise type collection unit 11 is specifically used to collect the category information of enterprises in the region. For example, it can classify enterprises into chemical enterprises, manufacturing enterprises, service enterprises, etc. This classification helps to identify potential pollution sources because different types of enterprises often have different pollution characteristics. In one embodiment of the present invention, the enterprise type collection unit 11 can obtain enterprise category information through questionnaire surveys, field visits, or database queries.

[0066] The enterprise production process collection unit 12 is responsible for collecting the enterprise production process. This step is crucial to understanding the specific pollution emissions of the enterprise. For example, for a chemical enterprise, its production process may include steps such as raw material mixing, reaction, separation, and refining, and each step may produce pollutants of different types and degrees. In actual applications, the enterprise production process collection unit 12 can obtain production process information through self-declaration by the enterprise, expert field visits, or remote monitoring.

[0067] The role of the regional planning processing unit 13 is to analyze the processes that affect regional environmental pollution sources based on enterprise categories and production processes. This unit integrates and analyzes the information collected by the first two units to identify key processes that may have an impact on the environment. For example, through analysis, it may be found that the reaction process of a chemical company is the main source of volatile organic compounds (VOCs) emissions, while the painting process of a manufacturing company is the main source of particulate matter emissions. This analysis provides targeted guidance for subsequent monitoring and early warning.

[0068] The air monitoring and early warning module 2 is one of the core components of the system, which is used to collect and analyze air quality data and issue early warnings. This module provides timely and accurate information support for environmental management decisions through real-time monitoring and intelligent analysis. Specifically, the air monitoring and early warning module 2 includes an air quality collection unit 21, an air quality analysis unit 22, and an air quality early warning unit 23.

[0069] The air quality collection unit 21 is responsible for collecting air quality parameters in the area in real time. These parameters usually include but are not limited to particulate matter (PM2.5, PM10), sulfur dioxide (SO2), nitrogen oxides (NO x ), ozone (O3), carbon monoxide (CO), etc. In a preferred embodiment of the present invention, the air quality collection unit 21 adopts a high-precision sensor network and is deployed in key locations in industrial parks, such as around enterprise emission outlets and near residential areas. These sensors can work 24 hours a day to ensure the continuity and real-time nature of the data.

[0070] The air quality analysis unit 22 is used to analyze the collected air quality parameters. This unit uses advanced data analysis techniques, such as machine learning algorithms, time series analysis, etc., to process and analyze massive amounts of air quality data. For example, it can identify abnormal fluctuations in air quality parameters, predict short-term air quality trends, analyze the correlation between different pollutants, etc. In practical applications, the air quality analysis unit 22 may use algorithms such as random forests and support vector machines to improve the accuracy and efficiency of the analysis.

[0071] The air quality warning unit 23 issues an environmental warning based on the analysis results. This unit sets a multi-level warning threshold, triggering a warning of the corresponding level according to the concentration level and duration of different pollutants. For example, when the PM2.5 concentration exceeds 75μg / m 3 If the concentration exceeds 150 μg / m 3 If the weather continues for more than 6 hours, an orange warning may be triggered. These thresholds are set based on national environmental protection standards and local conditions and can be adjusted as needed.

[0072] The main function of the campus communication management module 3 is to realize data interaction between components in the system. This module ensures the smooth flow of information throughout the system and is the key to the efficient operation of the system. In one embodiment of the present invention, the campus communication management module 3 adopts 5G network technology to ensure high-speed transmission and low-latency processing of large amounts of data. At the same time, the module also realizes encrypted transmission of data to protect the security of sensitive information.

[0073] Server 4 is the "brain" of the entire system, used to build environmental pollution source models and make predictions and warnings. Server 4 receives all data from other modules and uses big data analysis and artificial intelligence technology to build a complex environmental pollution source model. This model can not only reflect the current pollution situation, but also predict the pollution trend in the future. For example, by analyzing historical data and current meteorological conditions, Server 4 may predict that a certain area may experience severe smog weather in the next 48 hours, thereby issuing a warning in advance, allowing relevant departments enough time to take countermeasures.

[0074] The system of the present invention realizes comprehensive monitoring and intelligent early warning of environmental pollution in industrial parks through the collaborative work of various modules. It can not only detect environmental problems in a timely manner, but also provide targeted solutions, greatly improving the efficiency and accuracy of environmental management. For example, when the system detects that the concentration of VOCs emitted by a certain enterprise is abnormally high, it can immediately notify the environmental protection department to conduct on-site inspections, and at the same time send early warning information to the enterprise, suggesting that it check its production equipment or adjust its production process. This timely intervention can effectively prevent the occurrence or expansion of pollution incidents.

[0075] In general, the system of the present invention makes full use of advanced technologies such as the Internet of Things, big data and artificial intelligence to provide a comprehensive, intelligent and efficient solution for environmental protection in industrial parks. It can not only help managers better understand and control environmental pollution, but also provide important data support for policy formulation and long-term planning, thereby promoting the sustainable development of industrial parks.

[0076] In a preferred embodiment of the present invention, the air quality analysis unit 22 adopts an innovative multi-step analysis method to improve the accuracy and prediction ability of air quality parameter analysis. This method includes the following five steps: constructing a multidimensional gas concentration matrix, performing gas correlation analysis, constructing a gas diffusion group theory model, locating pollution sources, and generating a multi-level warning threshold matrix.

[0077] First, the method constructs a multidimensional gas concentration matrix. In practical applications, the air quality acquisition unit 21 may set up multiple sampling points in the industrial park, and each sampling point is equipped with a sensor array capable of detecting multiple gases. For example, in an industrial park covering an area of ​​10 square kilometers, there may be 100 sampling points evenly distributed, and each sampling point can detect 10 different gases at the same time. In this way, a multidimensional gas concentration matrix G containing a large amount of data is formed.

[0078] Preferably, the method for constructing the multidimensional gas concentration matrix G is as follows: an n-dimensional gas sensor array is set to collect data from m sampling points; and the collected data is organized into an m×n matrix G:

[0079]

[0080] Among them, g ij is the concentration value of the jth gas at the ith sampling point. For example, g 23 It may represent the concentration of sulfur dioxide (SO2) detected at the second sampling point, in μg / m 3 .

[0081] Next, the method of the present invention performs gas correlation analysis based on a multidimensional gas concentration matrix. The purpose of this step is to explore the relationship between different gas concentrations, which is crucial for understanding complex atmospheric chemical reactions and pollutant diffusion processes. In one embodiment of the present invention, the method of gas correlation analysis is as follows:

[0082] Based on the multidimensional gas concentration matrix G, the correlation matrix R between gases is calculated:

[0083]

[0084] Where m is the number of sampling points, 1 is an m-dimensional all-one column vector, G T represents the transposed matrix of G. This formula actually calculates the covariance matrix between different gas concentrations, which can reflect the correlation between the changes in different gas concentrations. For example, if the matrix R has r 12 The value of is close to 1, which means that the correlation between the first gas (denoted as g1) and the second gas (denoted as g2) is very high. Specifically, the correlation coefficient r 12 A value close to 1 indicates that the concentration changes of the two gases are almost completely synchronized, that is, their fluctuation trends in the time series are very similar. This may indicate that the concentration changes of the first gas and the second gas are highly positively correlated, which may suggest that they come from the same pollution source or participate in the same chemical reaction.

[0085] This means that when the concentration of one gas increases, the concentration of the other gas also increases almost simultaneously. When the concentration of one gas decreases, the concentration of the other gas also decreases almost simultaneously. The changes in the concentrations of the two gases are almost proportional. The two gases may come from the same pollution source or the same type of pollution source. For example, some industrial processes may emit multiple pollutants at the same time, resulting in highly correlated changes in the concentrations of these pollutants. The two gases may have similar physical and chemical properties, resulting in very similar transport, diffusion, and deposition behaviors in the atmosphere. Certain environmental factors (such as wind speed, temperature, humidity, etc.) may have similar effects on the changes in the concentrations of the two gases, thereby enhancing their correlation.

[0086] When constructing a gas diffusion group theory model, highly correlated gases may be considered to have similar diffusion patterns. This helps to simplify the model because it can be assumed that the spatial distribution and propagation behavior of the two gases are similar. When locating pollution sources, highly correlated gases can be considered together as a group of variables. If the concentration changes of two gases are highly consistent, then their pollution sources are likely to be located in the same location or very close. When generating a multi-level warning threshold matrix, highly correlated gases may share similar warning thresholds. This means that if the concentration of a gas exceeds the warning threshold, the other gas is also likely to approach or exceed its corresponding threshold, thereby triggering an alarm.

[0087] When performing multiple regression analysis or other statistical modeling, highly correlated variables may cause multicollinearity problems. This will affect the stability and explanatory power of the model. Therefore, when building an air quality analysis model, you may need to consider how to deal with this high correlation, such as reducing redundant information through principal component analysis (PCA) or other dimensionality reduction techniques.

[0088] After obtaining the gas correlation analysis results, the method further constructs a gas diffusion group theory model. This is an innovation of the present invention. Introducing the concept of group theory into the gas diffusion model can better describe the complex gas diffusion process. Specifically, in a preferred embodiment of the present invention, the method for constructing a gas diffusion group theory model includes:

[0089] Based on the correlation matrix R, construct the gas diffusion group D; define the operation of group D for:

[0090]

[0091] The elements of group D are the eigenvectors of the R matrix, representing the diffusion patterns of different gases. The definition of this group operation ensures that the result is still in the interval [0,1] and can be used to represent the normalized value of gas concentration. For example, if a represents the normalized concentration of SO2 in a certain place, and b represents the normalized concentration of NO x The normalized concentration of a○b can represent the comprehensive pollution degree after the two gases act together.

[0092] Based on the gas diffusion group theory model, the method further performs pollution source location. The purpose of this step is to infer the possible location of the pollution source based on the observed gas concentration distribution. In one embodiment of the present invention, the method for locating the pollution source is as follows:

[0093] Using the elements in group D, we construct the pollution source intensity function S(r,θ):

[0094]

[0095] Among them, Jl is the Bessel function of the first kind, α kl is its kth zero point, R′ is the radius of the considered area, c kl are unknown coefficients, r and θ represent radial and angular coordinates in polar coordinates, respectively. This function can describe the distribution of pollutants in space. For example, if the value of S(r,θ) is particularly large at a certain location (r,θ), then this location may be a pollution source.

[0096] Finally, the method generates a multi-level warning threshold matrix based on the pollution source location results. The purpose of this step is to set different levels of warning thresholds for different types of pollutants so as to issue an alarm in time. In a preferred embodiment of the present invention, the method for generating a multi-level warning threshold matrix includes:

[0097] Based on the pollution source intensity function S(r,θ), the warning threshold matrix T is constructed. * :

[0098]

[0099] Among them, q is the warning level number, which can usually be set to 3 to 5 levels, corresponding to different degrees of pollution such as light, moderate, and severe; p is the number of pollutant types, for example, it can include common pollutants such as PM2.5, PM10, SO2, NO2, and O3; t ij is the threshold value of the i-th level warning for the j-th pollutant. For example, t 23 It may indicate the concentration threshold of SO2 for moderate pollution warning, which may be set at 150 μg / m 3 .

[0100] Through the analysis of these five steps, the method of the present invention can comprehensively and deeply analyze the air quality data, not only identifying the current pollution situation, but also predicting the possible pollution trend in the future, providing strong support for environmental management decisions. For example, when the system detects that the SO2 concentration in a certain area is on the rise, and through the location of the pollution source, it is found that it may come from a chemical plant, the relevant departments can be notified in time to conduct on-site inspections to prevent further pollution.

[0101] In a preferred embodiment of the present invention, the pollution source location method is further refined to improve the accuracy and reliability of location. This method is based on the aforementioned gas diffusion group theory model and uses Bessel function expansion to describe the spatial distribution of pollution source intensity. Specifically, the method for constructing the pollution source intensity function S(r,θ) is as follows:

[0102]

[0103] In this formula, J lrepresents the Bessel function of the first kind, a special function that is particularly effective in describing wave phenomena in a circular region. kl It's J l The kth zero of is introduced to ensure the continuity of the function under boundary conditions. R′ represents the radius of the considered area. For example, in an industrial park with a radius of 5 kilometers, R′ may be 5000 meters. kl are unknown coefficients that can be determined by fitting actual observed data. r and θ represent the radial and angular coordinates in polar coordinates, respectively, so that the function can describe any point on a two-dimensional plane.

[0104] Bessel function of the first kind J l (x) is a special mathematical function that is widely used to describe wave phenomena in cylindrically symmetric systems. It is very useful in solving physical problems with circular or cylindrical symmetry, especially in the fields of acoustics, electromagnetism, and heat conduction. In this formula, J l It is used to describe the diffusion behavior of pollutants in a circular area, especially suitable for describing the distribution of pollutant concentration spreading outward from the center. kl It's J l The kth zero of (x), that is, satisfying J l (α kl )=0. Introducing α kl The purpose is to ensure the continuity and stability of the function under boundary conditions. Specifically, when r = R' (i.e., at the boundary of the considered region), will be equal to 0, thus ensuring that the function value is 0 at the boundary, which is consistent with the actual situation. R′ represents the radius of the considered area and defines the range of pollutant diffusion. For example, in an industrial park with a radius of 5 kilometers, R′ can be 5000 meters. By normalizing r to The formula can be applied to regions of different scales without rescaling the parameters. kl are coefficients determined by fitting actual observed data. These coefficients determine the weight of each Bessel function in the sum, reflecting the contribution of different modes to pollutant diffusion. kl , the model can better fit the actual observed pollutant concentration distribution, thereby improving the accuracy of the prediction. r and θ represent the radial and angular coordinates in polar coordinates, respectively, so that the function can describe any point on the two-dimensional plane. The radial coordinate r describes the distance that the pollutant diffuses outward from the center, while the angular coordinate θ describes the distribution of the pollutant in different directions. By introducing cos(lθ), the formula can capture the asymmetry of pollutants in different directions.

[0105] This formula is used to construct the pollution source intensity function S(r, θ) to help determine the location and intensity of the pollution source. By analyzing the spatial distribution of pollutant concentrations, the main pollution sources can be identified and their impact range can be assessed. Since the normalized radial coordinates are used in the formula It can be applied to areas of different scales, from city-level air pollution monitoring to local pollution source location in industrial areas. kl , the model can adapt to different environmental conditions and pollutant types, providing more flexible and accurate predictions. kl The continuity of the function at the boundary is ensured, unreasonable numerical fluctuations are avoided, and the model is made more stable and reliable.

[0106] The advantage of this method is that it can accurately describe the complex spatial distribution of pollutants. For example, if the value of S(r, θ) suddenly increases in a certain direction, it may indicate that there is a strong pollution source in that direction. By analyzing the peak position and shape of S(r, θ), environmental protection departments can quickly locate potential pollution sources and take targeted control measures.

[0107] Based on the location of the pollution source, the method of the present invention further generates a multi-level warning threshold matrix to achieve accurate pollution warning. In one embodiment of the present invention, the method for generating the multi-level warning threshold matrix T* is as follows:

[0108]

[0109] This matrix T * Each element t ij Represents the concentration threshold of the i-th level warning for the j-th pollutant. Preferably, the number of warning levels q can be set to 4, corresponding to the four levels of light, moderate, heavy and severe pollution. The number of pollutant types p can include common pollutants such as PM2.5, PM10, SO2, NO2, O3, etc., usually 5 to 7 types.

[0110] For example, t 23 The concentration threshold of SO2 may indicate a moderate pollution warning, which can be set to 150 μg / m according to national standards and local conditions. 3 When the monitored SO2 concentration exceeds this threshold, the system will trigger a moderate pollution warning. This multi-level warning mechanism can help relevant departments take corresponding response measures according to the severity of the pollution, such as issuing announcements and restricting the production of certain high-pollution industries.

[0111] Mild pollution (Level 1): When the pollutant concentration reaches the mild pollution threshold, the system will issue a prompt to remind relevant departments to strengthen monitoring and advise the public to take preventive measures such as reducing outdoor activities.

[0112] Moderate pollution (Level 2): ​​When the pollutant concentration reaches the moderate pollution threshold, the system will trigger a moderate pollution warning, and relevant departments may issue an announcement to advise sensitive groups (such as children, the elderly, and patients with respiratory diseases) to reduce going out and consider restricting production in certain high-pollution industries.

[0113] Severe pollution (Level 3): When the pollutant concentration reaches the severe pollution threshold, the system will issue a higher level of warning, and relevant departments may take more stringent measures, such as restricting the movement of motor vehicles and suspending construction.

[0114] Severe pollution (Level 4): When the pollutant concentration reaches the severe pollution threshold, the system will trigger the highest level of warning, and relevant departments may take emergency measures, such as closing schools, shutting down factories, and activating emergency plans.

[0115] Different pollutants have different impacts on human health and the environment, so setting independent thresholds for each pollutant can achieve more refined environmental management. For example, SO2 has a greater impact on the respiratory tract, while PM2.5 mainly affects lung health, so their warning thresholds may be different.

[0116] The warning threshold can be adjusted dynamically according to the season, weather conditions and regional characteristics. For example, during the winter heating season, SO2 emissions are usually higher, so the warning threshold for moderate pollution may need to be appropriately lowered to ensure timely response.

[0117] Through real-time monitoring and data analysis, the system can continuously optimize the warning threshold. For example, based on historical data and actual observation results, the c kl , and other parameters, so that the model can more accurately reflect the actual situation. This multi-level early warning mechanism can help relevant departments take corresponding countermeasures according to the severity of the pollution, such as issuing announcements, restricting the production of certain high-pollution industries, etc. By setting different levels of early warning thresholds, the system can respond to environmental pollution problems in a timely manner to ensure public health and environmental protection.

[0118] Finally, the present invention also provides an intelligent environmental monitoring and analysis method for environmental pollution sources based on the Internet of Things. This method integrates the functions of each module in the aforementioned system to form a complete workflow.

[0119] Specifically, the method comprises the following steps:

[0120] First, collect and process information about enterprises in the area. This step is completed by the Industrial Park Area Planning Module 1, which includes collecting information such as enterprise type and production process, and conducting preliminary analysis to identify potential pollution sources.

[0121] Secondly, collect and analyze air quality data. This step is performed by the air monitoring and early warning module 2, including real-time collection of air quality parameters and in-depth analysis using the aforementioned multi-step analysis method.

[0122] Next, the data exchange between the components in the system is realized. This step is the responsibility of the campus communication management module 3, which ensures the smooth transmission and exchange of data between the modules.

[0123] Finally, the environmental pollution source model is constructed and prediction and early warning are carried out. This step is completed by the server 4, which integrates the results of the previous steps, constructs a comprehensive environmental pollution source model, and performs pollution trend prediction and early warning based on this model.

[0124] It is worth noting that in the process of analyzing air quality data, this method uses the aforementioned innovative algorithm, including the steps of constructing a multidimensional gas concentration matrix, analyzing gas correlations, constructing a gas diffusion group theory model, locating pollution sources, and generating a multi-level warning threshold matrix. This comprehensive and in-depth analysis method enables the present invention to more accurately identify pollution sources, predict pollution trends, and issue warnings in a timely manner.

[0125] In general, the method provided by the present invention realizes comprehensive, real-time and intelligent monitoring of environmental pollution in industrial parks by integrating Internet of Things technology, big data analysis and artificial intelligence algorithms. It can not only detect and locate pollution sources in a timely manner, but also predict future pollution trends, providing strong support for environmental management decisions. For example, when the system predicts that severe pollution weather may occur in the next few days, relevant departments can take measures in advance, such as requiring high-polluting enterprises to limit or stop production, thereby effectively reducing the degree of pollution and protecting public health.

[0126] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto; any person familiar with the art who, within the scope disclosed by the present invention, makes substitutions or changes based on the scheme and improved concepts of the present invention shall be covered within the protection scope of the present invention.

Claims

1. The intelligent environmental monitoring and analysis system for environmental pollution sources based on the Internet of Things is characterized by: The system includes: Industrial park regional planning module, used to collect and process enterprise information in the area; Air monitoring and early warning module, used to collect and analyze air quality data and issue early warnings; The campus communication management module is used to realize data interaction between components in the system; Server, used to build environmental pollution source models and conduct predictions and early warnings.

2. The system according to claim 1, characterized in that The industrial park area planning module includes: Enterprise type collection unit, used to collect category information of enterprises in the area; Enterprise production process collection unit, used to collect enterprise production process; The regional planning processing unit is used to analyze the processes that affect regional environmental pollution sources based on enterprise categories and production processes.

3. The system according to claim 1, characterized in that The air monitoring and early warning module includes: Air quality collection unit, used to collect air quality parameters in the area in real time; An air quality analysis unit for analyzing air quality parameters; The air quality early warning unit is used to issue environmental early warnings based on the analysis results.

4. The system according to claim 3, characterized in that The air quality analysis unit performs air quality parameter analysis based on the following steps: Step A: construct a multidimensional gas concentration matrix; Step B: performing gas correlation analysis based on the multidimensional gas concentration matrix; Step C: constructing a gas diffusion group theory model based on the gas correlation analysis results; Step D: locating the pollution source based on the gas diffusion group theory model; Step E: Generate a multi-level warning threshold matrix based on the pollution source positioning result.

5. The system according to claim 4, characterized in that The method for constructing the multidimensional gas concentration matrix G in step A includes: Set up an n-dimensional gas sensor array to collect data from m sampling points; Organize the collected data into an m×n matrix G: Among them, g ij is the concentration value of the jth gas at the ith sampling point.

6. The system according to claim 4, characterized in that The method for performing gas correlation analysis in step B includes: calculating a correlation matrix R between gases based on the multidimensional gas concentration matrix G: Where m is the number of sampling points, 1 is an m-dimensional all-one column vector, G T Represents the transposed matrix of G.

7. The system according to claim 4, characterized in that The method for constructing the gas diffusion group theory model in step C includes: Based on the correlation matrix R, a gas diffusion group D is constructed; and the operation of group D is defined. for The elements of group D are the eigenvectors of the R matrix, representing the diffusion patterns of different gases.

8. The system according to claim 4, characterized in that The method for locating the pollution source in step D includes: using the elements in the group D to construct a pollution source intensity function S(r,θ): Among them, J l is the Bessel function of the first kind, α kl is its kth zero point, R′ is the radius of the considered area, c kl are unknown coefficients, r and θ represent the radial and angular coordinates in polar coordinates, respectively.

9. The system according to claim 4, characterized in that The method for generating a multi-level warning threshold matrix in step E comprises: Based on the pollution source intensity function S(r,θ), the warning threshold matrix T is constructed. * : Among them, q is the number of warning levels, p is the number of pollutant types, and t i j is the threshold value of the i-th level warning for the j-th pollutant.

10. An intelligent environmental monitoring and analysis method for environmental pollution sources based on the Internet of Things, characterized in that: The method includes: Collect and process enterprise information in the region; Collect and analyze air quality data; Realize data interaction between components in the system; Construct environmental pollution source models and conduct prediction and early warning; The air quality data analysis process includes the steps described in any one of claims 4 to 9.

Images