A pollution early warning detection system based on big data

Abstract

The invention relates to the field of pollution detection, and discloses a pollution early warning detection system based on big data, comprising an environmental data acquisition module for real-time collection of pollutant concentration and meteorological parameters of each monitoring area, an environmental data processing module for obtaining pollutant concentration indicators by inputting pollutant concentration into a pre-trained neural network model; a pollution analysis module for calculating according to the pollutant concentration indicators and meteorological parameters of each monitoring area to obtain a pollution early warning coefficient, and comparing the pollution early warning coefficient with a preset pollution early warning interval, marking a monitoring area exceeding the pollution early warning value as a polluted area, marking a monitoring area in the pollution early warning interval as a critical area, and performing pollution prediction analysis; marking a monitoring area below the pollution early warning interval as a safe area; and a pollution early warning module for judging whether to issue an early warning and determining a corresponding early warning level according to the comparison result of the pollution analysis module.

Description

Technical Field

[0001] The present invention relates to the field of pollution detection, and in particular to a pollution early warning detection system based on big data. Background Art

[0002] Environmental pollution is becoming increasingly serious, posing a serious threat to the ecological environment and human health. Therefore, environmental monitoring has become an important part of environmental protection. Traditional environmental monitoring methods often have problems such as incomplete data collection, untimely processing, and inaccurate early warning.

[0003] Therefore, the existing environmental pollution detection system has introduced big data technology. Big data technology refers to providing targeted information services based on massive, high-speed and diversified data through innovative data processing and analysis technologies. With the continuous advancement of information technology, big data technology has penetrated into various industries, including the field of environmental monitoring. The application of big data technology makes the collection, processing, analysis and early warning of environmental monitoring data more efficient and accurate.

[0004] However, in actual applications, it was found that the existing environmental pollution detection system still has some defects. For example, if only a single indicator related to pollutant concentration is detected during the environmental pollution detection process, there will be human interference in the pollutant concentration-related indicators, causing data fraud, and ultimately leading to inaccurate detection results and missed or false pollution warnings. Summary of the invention

[0005] The purpose of the present invention is to provide a pollution early warning detection system based on big data, so as to solve the technical problem that only detecting the relevant indicators of pollutant concentration will lead to human interference in the relevant indicators of pollutant concentration, causing data fraud, and ultimately leading to inaccurate detection results, omissions or false alarms of pollution early warnings.

[0006] The purpose of the present invention can be achieved through the following technical solutions:

[0007] A pollution early warning detection system based on big data, comprising:

[0008] The environmental data acquisition module is divided into multiple monitoring areas according to the factory area. Each monitoring area is equipped with a group of monitoring points. Each monitoring point is equipped with optical sensors, electrochemical sensors and meteorological sensors to collect pollutant concentrations and meteorological parameters of each monitoring area in real time and store them in the database;

[0009] Environmental data processing module, which obtains pollutant concentration indicators by inputting pollutant concentrations into a pre-trained neural network model;

[0010] The pollution analysis module is used to calculate the pollution warning coefficient based on the pollutant concentration index and meteorological parameters of each monitoring area, and compare the pollution warning coefficient with the preset pollution warning interval;

[0011] Mark the monitoring area that exceeds the pollution warning value as a polluted area;

[0012] Mark the monitoring area in the pollution warning range as a critical area and conduct pollution prediction analysis;

[0013] Mark the monitoring area below the pollution warning range as a safe area;

[0014] The pollution warning module issues a severe warning for the polluted area based on the comparison results of the pollution analysis module, and determines whether to issue a general warning for the critical area based on the results of the pollution prediction analysis, and no warning is issued for the safe area.

[0015] Among them, the pollutant concentration parameters include PM2.5, PM10, sulfur dioxide and nitrogen oxides, and the meteorological parameters include temperature T, humidity RG and wind speed WS.

[0016] A further solution is to obtain the pollution warning coefficient as follows:

[0017] Substitute the pollutant concentration index and temperature T, humidity RG and wind speed WS of each monitoring area into the formula:

[0018] In the calculation, we get Monitoring area Pollution warning coefficient at all times ;

[0019] In the formula, , is the preset weight factor, For the The total number of monitoring points in the monitoring area, For the Monitoring points, For the The deviation coefficient of the pollutant concentration index in each monitoring area during the previous detection period, is the pollution diffusion function regarding temperature, humidity and wind speed, For the Within the monitoring area The pollution diffusion value at the time, For the Within the monitoring area Pollutant concentration index at a given moment.

[0020] In a further embodiment, the Deviation coefficient of pollutant concentration index in each monitoring area during the previous detection period The acquisition process is:

[0021] The current time is the end point of the detection period. , at preset intervals Determine the starting point of the detection period for the time span ;

[0022] Get in Curve of internal pollutant concentration index changing with time ;

[0023] By formula Calculate the first Deviation coefficient of pollutant concentration index in each monitoring area during the previous detection period ;

[0024] In the formula , They are The maximum and minimum values ​​of the internal pollutant concentration index, , is the preset scale factor, It is a reference curve of pollutant concentration indicators changing over time based on big data.

[0025] Among them, the pollution diffusion function about temperature, humidity and wind speed ;

[0026] In the formula, , , is the weight coefficient, is the exponential coefficient, is a constant.

[0027] In a further embodiment, the process of pollution prediction analysis is as follows:

[0028] The critical area is divided into Sort in descending order;

[0029] Pass the formula for each critical section in sequence:

[0030] Calculate the pollution trend coefficient ;

[0031] when When , it is judged that the current critical area has a potential pollution risk, otherwise, it is judged that the current critical area does not have a potential pollution risk; is the preset pollution trend threshold;

[0032] In the formula, For the Real-time values ​​of pollutant concentration parameters, For the Reference values ​​of pollutant concentration parameters, is the standard value, is the judgment function, when >0 o'clock, ,otherwise, , For the The coefficient of variation of the pollutant concentration is , is the number of items of pollutant concentration parameter, is the meteorological coefficient of variation.

[0033] A further solution is described ; In the formula, No. The value of the sampling time point, is the total number of sampling time points, For the The mean of the pollutant concentration parameters.

[0034] In a further embodiment, the meteorological coefficient of variation The acquisition process is:

[0035] Obtain a curve of temperature variation over time in each critical area within a detection period , Curve of humidity changing with time , Curve of wind speed changing with time .

[0036] By formula Calculate the meteorological variation coefficient of the current critical area ,in is the reference wind speed change, is the reference temperature change, is the reference wind speed change; , , The default impact factor.

[0037] Beneficial effects of the present invention:

[0038] (1) The present invention stores the detected data in the system database for sharing through the environmental data acquisition module, and the environmental data processing module retrieves the data from the database and processes it to obtain the pollutant concentration index, and then uses the pollution analysis module to conduct a comprehensive analysis of the pollutant concentration index and meteorological parameters in each monitoring area to establish a pollution warning model, so that the accuracy and response speed of the pollution warning are effectively improved; at the same time, the results of the comprehensive analysis can be directly reflected through the pollution warning coefficient, and then the pollution warning coefficient is compared with the preset pollution warning interval. According to the comparison result, each monitoring area is accurately divided into a polluted area, a critical area and a safe area, thereby realizing the classification of the pollution conditions in different monitoring areas;

[0039] (2) The present invention can associate temperature, humidity and wind speed in meteorological parameters with pollutant concentration indicators, thereby establishing a pollution prediction model. Compared with a single detection of pollutant concentration indicators, the influence of temperature, humidity and wind speed on the pollutant concentration indicators can be incorporated into the model, thereby effectively improving the accuracy of pollution prediction. At the same time, it can prevent human interference in parameters related to pollutant concentration during the monitoring and maintenance process, which would lead to inaccurate pollution prediction results. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] The present invention will be further described below in conjunction with the accompanying drawings.

[0041] Figure 1 is a system structure block diagram of the present invention;

[0042] Figure 2 It is a flow chart of pollution prediction analysis in the present invention. DETAILED DESCRIPTION

[0043] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0044] See also Figure 1-Figure 2 As shown, the present invention is a pollution early warning detection system based on big data, comprising:

[0045] The environmental data acquisition module is divided into multiple monitoring areas according to the factory area. Each monitoring area is equipped with a group of monitoring points. Each monitoring point is equipped with optical sensors, electrochemical sensors and meteorological sensors to collect pollutant concentrations and meteorological parameters of each monitoring area in real time and store them in the database; wherein the pollutant concentration parameters include PM2.5, PM10, sulfur dioxide and nitrogen oxides, and the meteorological parameters include temperature T, humidity RG and wind speed WS;

[0046] Environmental data processing module, which obtains pollutant concentration indicators by inputting pollutant concentrations into a pre-trained neural network model;

[0047] The pollution analysis module is used to calculate the pollution warning coefficient based on the pollutant concentration index and meteorological parameters of each monitoring area, and compare the pollution warning coefficient with the preset pollution warning interval;

[0048] Mark the monitoring area that exceeds the pollution warning value as a polluted area;

[0049] Mark the monitoring area in the pollution warning range as a critical area and conduct pollution prediction analysis;

[0050] Mark the monitoring area below the pollution warning range as a safe area;

[0051] The pollution warning module issues a severe warning for the polluted area based on the comparison results of the pollution analysis module, and determines whether to issue a general warning for the critical area based on the results of the pollution prediction analysis, and no warning is issued for the safe area.

[0052] The present invention stores the detected data in the database of the system for sharing through the environmental data acquisition module, and the environmental data processing module retrieves the data from the database and processes it to obtain the pollutant concentration index, and then uses the pollution analysis module to conduct a comprehensive analysis of the pollutant concentration index and meteorological parameters in each monitoring area to establish a pollution warning model, so that the accuracy and response speed of the pollution warning are effectively improved; at the same time, the result of the comprehensive analysis can be directly reflected through the pollution warning coefficient, and then the pollution warning coefficient is compared with the preset pollution warning interval, and each monitoring area is accurately divided into a polluted area, a critical area and a safe area according to the comparison result, so as to realize the classification of the pollution conditions in different monitoring areas;

[0053] For the polluted area, a severe warning is directly issued, indicating that there is a high probability of a pollution source in this area, so timely response treatment is required. For the critical area, although the pollution has not directly exceeded the standard, in order to predict the pollution situation in advance and prevent the pollution situation in the critical area from worsening, more resources need to be invested. Therefore, the critical area is predicted by pollution prediction analysis. Obviously, if there is a potential pollution risk, a general warning is issued. After the pollution area with a severe warning is handled, the critical area with potential pollution risk is pre-processed to reduce the occurrence of pollution deterioration.

[0054] It should be noted that the working principles of optical sensors, electrochemical sensors and meteorological sensors are existing technologies, and the specific installation method is selected according to the actual situation on site, so there is no need to elaborate here.

[0055] The process of obtaining the pollution warning coefficient is:

[0056] Substitute the pollutant concentration index and temperature T, humidity RG and wind speed WS of each monitoring area into the formula:

[0057] In the calculation, we get Monitoring area Pollution warning coefficient at all times ;

[0058] In the formula, , is a preset weight factor, selected and determined based on historical data. For the The total number of monitoring points in the monitoring area, For the Monitoring points, For the The deviation coefficient of the pollutant concentration index in each monitoring area during the previous detection period, is the pollution diffusion function regarding temperature, humidity and wind speed, For the Within the monitoring area The pollution diffusion value at the time, For the Within the monitoring area Pollutant concentration index at a given moment.

[0059] The said Deviation coefficient of pollutant concentration index in each monitoring area during the previous detection period The acquisition process is:

[0060] The current time is the end point of the detection period , at preset intervals Determine the starting point of the detection period for the time span ;

[0061] Get in Curve of internal pollutant concentration index changing with time ;

[0062] By formula Calculate the first Deviation coefficient of pollutant concentration index in each monitoring area during the previous detection period ;

[0063] In the formula , They are The maximum and minimum values ​​of the internal pollutant concentration index, , It is a preset proportional coefficient, which is determined based on historical data and empirical data. It is a reference curve of pollutant concentration indicators changing over time based on big data.

[0064] Among them, the pollution diffusion function about temperature, humidity and wind speed ;

[0065] In the formula, , , is the weight coefficient, is the exponential coefficient, is a constant; , , , , Determined by fitting historical data and experimental data.

[0066] The present invention provides a method for obtaining a pollution warning coefficient, first by substituting the pollutant concentration index and the temperature T, humidity RG and wind speed WS of each monitoring area into the formula: In the calculation, we get Monitoring area Pollution warning coefficient at all times ,in, , ; Through the above formula group, the temperature, humidity and wind speed in the meteorological parameters can be associated with the pollutant concentration index, so as to establish a pollution prediction model. Compared with the single detection of pollutant concentration index, the influence of temperature, humidity and wind speed on the pollutant concentration index can be included, so as to effectively improve the accuracy of pollution prediction. At the same time, it can prevent human interference with the parameters related to pollutant concentration during the monitoring and operation process, which will lead to inaccurate pollution prediction results;

[0067] It should be noted that the deviation coefficient of the pollutant concentration index in the previous detection period The larger the value, the higher the pollutant concentration index. The greater the cumulative change in the previous detection period, the greater the probability of contamination. It indicates the difference between the maximum and minimum values ​​in the previous detection period. Obviously, if The smaller the value, the smaller the overall change of the pollutant concentration index is, so the probability of the existence of a pollution source is greater. The larger the value is, the greater the change in the overall image of the pollutant concentration index is. Therefore, the probability of pollution caused by the pollution source near the monitoring area spreading to the area is high. Through the above process, it can be judged whether there is pollution in the monitoring area and whether the pollution is caused by the pollution source, which provides strong support for subsequent judgment and processing, so as to further improve the accuracy of the pollution prediction model.

[0068] At the same time, through the function It can be seen that As a constant term, it provides a baseline or offset for the entire function. When all other variables T, RH, and WS are zero, the function will not output zero, but output The value of can also be adjusted during the model building process to minimize the difference between the model prediction value and the actual observation value to achieve the purpose of compensating the model error; Determines the nonlinear influence of wind speed WS on the function output value. , then the effect of wind speed is linear; if or , the influence of wind speed is nonlinear; through the above process, a prediction model of temperature, humidity and wind speed on pollutant diffusion is established. Obviously, the larger the function output value, the faster the pollutant diffuses, and vice versa.

[0069] The process of pollution prediction analysis is as follows:

[0070] The critical area is divided into Sort in descending order;

[0071] Pass the formula for each critical section in sequence:

[0072] Calculate the pollution trend coefficient ;

[0073] when When , it is judged that the current critical area has a potential pollution risk, otherwise, it is judged that the current critical area does not have a potential pollution risk; is the preset pollution trend threshold;

[0074] In the formula, For the Real-time values ​​of pollutant concentration parameters, For the Reference values ​​of pollutant concentration parameters, is the standard value, is the judgment function, when hour, ,otherwise, , For the The coefficient of variation of the pollutant concentration is , is the number of items of pollutant concentration parameter, is the meteorological coefficient of variation.

[0075] Said ; In the formula, No. The value of the sampling time point, is the total number of sampling time points, For the The mean of the pollutant concentration parameters.

[0076] The present invention provides a method for predicting and analyzing the pollution of a critical area. Specifically, the formula Calculate the pollution trend coefficient , and then the calculated pollution trend coefficient With the preset pollution trend threshold For comparison, if When , it is judged that the current critical area has a potential pollution risk, otherwise, it is judged that the current critical area does not have a potential pollution risk; Obviously, when The larger the value, the greater the fluctuation of the pollutant concentration in the qth item, so the probability of the pollution situation deteriorating is higher, so the pollution trend coefficient The larger the standard deviation, the calculate , so as to achieve the purpose of accurately obtaining the concentration parameters of each pollutant, improve the accuracy of pollution prediction analysis, and prevent false alarms; judgment function The larger the value, the greater the difference between the current pollutant concentration and the reference value, so the higher the probability of the pollution situation worsening. The concentration of each pollutant is summed up in a cumulative manner to obtain the overall pollution trend coefficient. , it is clear that the pollution trend coefficient The larger it is, the higher the probability of the pollution situation deteriorating. Through the above process, the potential pollution risk of the critical area can be effectively judged and misjudgment can be prevented.

[0077] The meteorological coefficient of variation The acquisition process is:

[0078] Obtain a curve of temperature variation over time in each critical area within a detection period , Curve of humidity changing with time , Curve of wind speed changing with time .

[0079] By formula Calculate the meteorological variation coefficient of the current critical area ,in is the reference wind speed change, is the reference temperature change, is the reference wind speed change; , , The default impact factor.

[0080] The present invention comprehensively analyzes the changes in wind speed, temperature, and humidity during the detection period to establish an overall meteorological change model, which reflects the impact of the current critical zone meteorological changes on the potential pollution risk, thereby achieving a more accurate prediction of the potential pollution risk and achieving a timely response to the pollution situation in the critical zone; through the formula It can be seen that The smaller, The smaller the temperature and wind speed change during the test period, the smaller the meteorological variation coefficient The smaller, The larger the value is, the greater the humidity change during the detection period is, so the meteorological variation coefficient The larger the wind speed, the faster the pollutants will diffuse. The higher the temperature, the more molecular motion will intensify, which will also cause the pollutants to diffuse faster. The higher the humidity, the more pollutant particles will be absorbed, and the slower the pollutants will diffuse. On the contrary, the lower the humidity, the faster the pollutants will diffuse. Therefore, the meteorological variation coefficient The smaller it is, the greater the impact on the pollutant concentration forecast of the monitored area, and the greater the impact on the meteorological variation coefficient Dynamic adjustment makes pollution prediction and analysis more accurate.

[0081] It should be noted that the calculation formulas and various parameters involved in the calculation in the present invention have been dimensionally processed in advance, and the process of dimensionless processing is well known in the industry and will not be described here.

[0082] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A pollution early warning detection system based on big data, characterized in that: include: The environmental data acquisition module is divided into multiple monitoring areas according to the factory area. Each monitoring area is equipped with a group of monitoring points. Each monitoring point is equipped with optical sensors, electrochemical sensors and meteorological sensors to collect pollutant concentrations and meteorological parameters of each monitoring area in real time and store them in the database; Environmental data processing module, which obtains pollutant concentration indicators by inputting pollutant concentrations into a pre-trained neural network model; The pollution analysis module is used to calculate the pollution warning coefficient based on the pollutant concentration index and meteorological parameters of each monitoring area, and compare the pollution warning coefficient with the preset pollution warning interval; Mark the monitoring area that exceeds the pollution warning value as a polluted area; Mark the monitoring area in the pollution warning range as a critical area and conduct pollution prediction analysis; Mark the monitoring area below the pollution warning range as a safe area; The pollution warning module issues a severe warning for the polluted area based on the comparison results of the pollution analysis module, and determines whether to issue a general warning for the critical area based on the results of the pollution prediction analysis, and does not issue a warning for the safe area; The process of calculating the pollution warning coefficient based on the pollutant concentration index and meteorological parameters of each monitoring area is as follows: Substitute the pollutant concentration index and temperature T, humidity RG and wind speed WS of each monitoring area into the formula: In the calculation, we get Monitoring area Pollution warning coefficient at all times ; In the formula, , is the preset weight factor, For the The total number of monitoring points in the monitoring area, For the Monitoring points, For the The deviation coefficient of the pollutant concentration index in each monitoring area during the previous detection period, is the pollution diffusion function regarding temperature, humidity and wind speed, For the Within the monitoring area The pollution diffusion value at the time, For the Within the monitoring area Pollutant concentration index at the moment; The pollution diffusion function for temperature, humidity, and wind speed ; In the formula, , , is the weight coefficient, is the exponential coefficient, is a constant; The process of pollution prediction analysis is as follows: The critical area is divided into Sort in descending order; Pass the formula for each critical section in sequence: Calculate the pollution trend coefficient ; when When , it is judged that the current critical area has a potential pollution risk, otherwise, it is judged that the current critical area does not have a potential pollution risk; is the preset pollution trend threshold; In the formula, For the Real-time values ​​of pollutant concentration parameters, For the Reference values ​​of pollutant concentration parameters, is the standard value, is the judgment function, when >0 o'clock, ,otherwise, , For the The coefficient of variation of the pollutant concentration is ∈[1, ], is the number of items of pollutant concentration parameter, is the meteorological variation coefficient; Said ; In the formula, No. The value of the sampling time point, is the total number of sampling time points, For the The mean of the pollutant concentration parameters; The meteorological coefficient of variation The acquisition process is: Obtain a curve of the temperature change over time in each critical area within a detection period , Curve of humidity changing with time , Curve of wind speed changing with time ; By formula Calculate the meteorological variation coefficient of the current critical area ,in is the reference wind speed change, is the reference temperature change, is the reference wind speed change; , , is the preset impact factor; The said Deviation coefficient of pollutant concentration index in each monitoring area during the previous detection period The acquisition process is: The current time is the end point of the detection period , at preset intervals Determine the starting point of the detection period for the time span ; Get in Curve of internal pollutant concentration index changing with time ; By formula Calculate the first Deviation coefficient of pollutant concentration index in each monitoring area during the previous detection period ; In the formula , They are The maximum and minimum values ​​of the internal pollutant concentration index, , is the preset scale factor, It is a reference curve of pollutant concentration indicators changing over time based on big data.

2. The pollution early warning detection system based on big data according to claim 1 is characterized in that: The pollutant concentration parameters include PM2.5, PM10, sulfur dioxide and nitrogen oxides, and the meteorological parameters include temperature T, humidity RG and wind speed WS.

Images