Intelligent environment monitoring management system based on big data

Abstract

The invention discloses an intelligent environment monitoring management system based on big data, and relates to the technical field of environment monitoring management, and the system specifically comprises an environment change data collection module, an objective function setting module, and an environment prediction management module, the multi-source heterogeneous data collection module is used for collecting multi-source heterogeneous data of an environment monitoring area in real time, the target function setting module is used for establishing an environment monitoring management target function of the environment monitoring area and setting constraint conditions of the environment monitoring area based on historical work intensity data of an enterprise factory, and the environment prediction management module is used for predicting the environment. According to the scheme, an environment change data prediction model is established based on historical environment change data of an enterprise factory environment monitoring area, environment monitoring control of an enterprise factory is realized by predicting the environment change data, and digital management considering environment protection and production benefits is realized through high-precision perception-multi-target optimization-intelligent prediction.

Description

Technical Field

[0001] The present invention belongs to the technical field of environmental monitoring and management, and particularly relates to an intelligent environmental monitoring and management system based on big data. Background Art

[0002] Existing monitoring systems often rely on discrete point monitoring, and there is a lack of correlation and integrity between data, making it difficult to reflect the overall pollution diffusion law of the region. This isolated data state makes it difficult for environmental management departments to comprehensively grasp the pollution situation and unable to conduct effective pollution tracking and source investigation.

[0003] Secondly, response lag is also a significant problem in environmental monitoring systems. Due to delays in the processes of data collection, transmission, and processing, environmental management departments often miss the best intervention opportunity after receiving pollution data. This lagged response not only increases the difficulty of pollution control but also may cause greater damage to the ecological environment and residents' health.

[0004] Furthermore, extensive regulation is another pain point in current environmental monitoring systems. Existing environmental management systems often lack dynamic coupling with production activities and are difficult to balance environmental protection and production capacity requirements. This extensive regulation method may not only lead to waste of environmental resources but also cause unnecessary interference to the normal production activities of enterprises.

[0005] Therefore, there is an urgent need for an intelligent environmental monitoring and management system based on big data to solve the above problems. Summary of the Invention

[0006] The purpose of the present invention is to provide an intelligent environmental monitoring and management system based on big data, which is used to solve the technical problems of data isolation, response lag, extensive regulation, significant environmental interference, and insufficient prediction ability existing in the prior art.

[0007] To achieve the above purpose, the present invention adopts the following technical solutions:

[0008] An intelligent environmental monitoring and management system based on big data, comprising:

[0009] An environmental change data collection module, which is used to determine a specific environmental monitoring area and collect multi-source heterogeneous data of the environmental monitoring area in real time;

[0010] A target function setting module, which is used to establish an environmental monitoring and management target function for the environmental monitoring area and set constraint conditions for the environmental monitoring area based on the historical work intensity data of enterprise factories;

[0011] The environmental prediction management module establishes an environmental change data prediction model based on the historical environmental change data of the environmental monitoring area of the enterprise factory, and realizes the environmental monitoring and control of the enterprise factory by predicting the environmental change data.

[0012] Furthermore, determine the specific environmental monitoring area and collect multi-source heterogeneous data of the environmental monitoring area in real time. The specific method is as follows:

[0013] Identify the enterprise factory and its surrounding sensitive areas, delimit the environmental monitoring area, construct a three-dimensional monitoring network at the environmental monitoring area, divide the time points by t duration, and collect multi-source heterogeneous data of the environmental monitoring area at each time point. The specific method is to establish a ground sensing layer by deploying a temperature and humidity adaptive compensation sensor array, collect the change data of PM2.5 and sulfur dioxide concentrations in each time period within the environmental monitoring area, divide the monitoring area into A×A grids, deploy 1 sensor node at the center of each grid, and each sensor node contains a PM2.5 sensor and a sulfur dioxide sensor. Encrypt the boundary area to a spacing, construct a temperature and humidity compensation formula, and process the output value of the sensor through the temperature and humidity compensation formula;

[0014] By accessing the NASA Earthdata real-time data stream, collect the Landsat-9 multispectral data of the environmental monitoring area in real time. By analyzing the Landsat-9 multispectral data, construct a normalized difference vegetation index using the red light band and the near-infrared band, divide the time period by T duration, and compare the normalized difference vegetation indices of the same time period in different historical time periods.

[0015] Furthermore, construct a temperature and humidity compensation formula. The specific method is as follows:

[0016] Use S(x) = s(x) × [1 + a1×(H(x) - h) + a2×(R(x) - r)] to represent the temperature and humidity compensation formula, where x represents the xth time point, S(x) represents the sensor output value after temperature and humidity compensation at the xth time point, s(x) represents the original sensor output value at the xth time point, H(x) represents the temperature value at the xth time point, h represents the standard temperature, R(x) represents the humidity value at the xth time point, r represents the standard humidity, a1 is the temperature compensation coefficient, and a2 is the humidity compensation coefficient.

[0017] Furthermore, establish an environmental monitoring management objective function for the environmental monitoring area. The specific method is as follows:

[0018] Collect the environmental change data of each environmental monitoring area in N time periods. Use a unique number to identify each environmental monitoring area in the enterprise factory, and each number corresponds to only one environmental monitoring area, obtaining a set of serial numbers representing each area of the enterprise factory, denoted as {1,..., i,..., n}, where i represents the i-th environmental monitoring area and n represents that there are n environmental monitoring areas in the enterprise factory. Use the formula represents the environmental monitoring management objective function, where i represents the i-th area, so(i) represents the average sulfur dioxide concentration in the i-th area within a time period, Δpm(i) represents the number of time periods when the PM2.5 concentration in the i-th area decreases within a time period, nv(i) represents the average value of the normalized difference vegetation index in the i-th area within a time period, Δfg(i) represents the number of time periods when the normalized difference vegetation index in the i-th area increases within a time period, g(i) represents the pollution weight coefficient of the i-th area, and h(i) represents the greening weight coefficient of the i-th area.

[0019] Furthermore, the number of time periods when the PM2.5 concentration decreases includes:

[0020] Use the formula to represent the number of time periods when the PM2.5 concentration decreases, where i represents the i-th area, x represents the x-th time point, N represents the total number of time points within a time period, pm(i, x) represents the PM2.5 concentration value collected at the x-th time point in area i, and H(*) represents the step function. When * is less than or equal to 0, H(*) is equal to 0, and when * is greater than 0, H(*) is equal to 1.

[0021] Furthermore, the specific method for the number of time periods when the normalized difference vegetation index increases is:

[0022] Use the formula to represent the number of time periods when the normalized difference vegetation index increases. Here, i represents the i-th area, x represents the x-th time point, N represents the total number of time points within a time period, nv(i, x) represents the normalized difference vegetation index collected at the x-th time point in area i, and H(*) represents the step function. When * is less than or equal to 0, H(*) is equal to 0, and when * is greater than 0, H(*) is equal to 1.

[0023] Furthermore, the specific method for setting the constraint conditions of the environmental monitoring area is:

[0024] Use the inequality to represent the equipment operation requirement of area i, where m represents the m-th time period, M represents the total number of time periods within a time period, zx(m) represents the equipment online rate at the m-th time period, G is the online rate threshold, I[*] is the conditional function, when the condition * is satisfied, I[*] is equal to 1, and A is the equipment operation requirement threshold;

[0025] Using inequalities to represent the benchmark demand for production capacity in region i, where m represents the m-th time period, M represents the total number of time periods in a time cycle, Q(i,m) represents the actual production capacity of region i in the m-th time period, and Q(i) represents the benchmark value of the production capacity of the i-th region, represents the production capacity weight coefficient of the i-th region, and a is a constant;

[0026] Using inequalities to represent the energy consumption intensity demand of region i, where E(i,m) represents the total energy consumption of region i in the m-th time period, GD(i,m) represents the total economic output value of region i in the m-th time period, E(i) represents the benchmark energy consumption intensity, i.e., the energy consumption standard per unit of GDP, ΔT represents the difference between the average temperature of the working equipment in region i and the benchmark temperature in a time cycle, β(i) represents the energy consumption weight coefficient of region i, and b is the temperature influence coefficient.

[0027] Furthermore, an environmental change data prediction model is established. The specific method is as follows:

[0028] Extract historical environmental change data, establish an environmental change data prediction model based on the historical environmental change data of consecutive historical time periods, and use the formula yc(i,m) = y0 + y1×[y(i,m - 1)] 2 +…+ y k ×y(i,m - 1)] k to represent the environmental change data prediction model, where m represents the m-th time period, i represents the i-th region, y0, y1, y k represent model parameters, k represents the model order. By inputting the historical environmental change data into the environmental change data prediction model, a predicted value is obtained. Based on the true value of the energy change data for the corresponding time period, the target squared error between the predicted value and the true value is determined. Set the target squared error threshold J. When the target squared error between the predicted value and the true value is less than or equal to the threshold J, it is determined that the training of the environmental change data prediction model is completed. When the target squared error between the predicted value and the true value is greater than the threshold J, it is determined that the training of the environmental change data prediction model is not completed. By adjusting the parameter values and the order of the model, the target squared error between the predicted value and the true value is reduced.

[0029] Furthermore, by predicting environmental change data, environmental monitoring and control of enterprise factories are realized. The specific method is as follows:

[0030] Use the trained prediction model to generate predicted values for the next W time periods based on the environmental change data in the current time period. Set the pollution ratio threshold O, and determine the time periods during which the PM2.5 concentration or sulfur dioxide concentration increases in the predicted future time period, which are recorded as pollution time periods. If the ratio of the number of pollution time periods to W is greater than or equal to the threshold O, it is determined that the current pollution is severe, and emission reduction measures or production plan adjustments need to be taken.

[0031] In summary, due to the adoption of the above technical solutions, the beneficial effects of the present invention are as follows:

[0032] 1. By determining specific environmental monitoring areas and collecting multi-source heterogeneous data in these areas in real time, the present invention can comprehensively and accurately grasp the environmental conditions of enterprise factories and surrounding sensitive areas. The use of a temperature and humidity adaptive compensation sensor array effectively reduces the interference of temperature and humidity on sensor measurements and improves the accuracy of data collection. The deployed ground sensing layer and grid-based sensor node layout ensure the refinement and full coverage of environmental monitoring.

[0033] 2. By establishing an environmental monitoring management objective function and setting constraint conditions based on the historical work intensity data of enterprise factories, the present invention realizes the scientific management of the environmental resource consumption in various regions of enterprise factories. Multiple indicators such as the sulfur dioxide concentration, the number of time periods with a decrease in the PM2.5 concentration, the normalized difference vegetation index, and the number of time periods with its increase, as well as the pollution weight coefficient and greening weight coefficient of different regions, are comprehensively considered in the objective function, making the environmental monitoring management more comprehensive and detailed.

[0034] 3. The environmental change data prediction model established based on historical environmental change data in the present invention can predict the environmental change data in future time periods, providing strong support for the environmental monitoring and control of enterprise factories. By adjusting the parameter values and orders of the prediction model, the performance of the prediction model can be continuously optimized, and the prediction accuracy can be improved. According to the prediction results, enterprise factories can timely discover potential pollution problems and take corresponding emission reduction measures or adjust production plans to cope with environmental pollution risks. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or in the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention, and those of ordinary skill in the art can obtain other drawings without creative efforts based on these drawings.

[0036] Figure 1 Shows a module diagram of an intelligent environmental monitoring management system based on big data;

[0037] Figure 2Shows the method step diagram of intelligent environmental monitoring and management based on big data;

[0038] Figure 3 Shows the method step diagram of environmental monitoring and control in an enterprise factory. Detailed implementation manner

[0039] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0040] As Figure 1 The intelligent environmental monitoring and management system based on big data shown specifically includes the following steps:

[0041] The environmental change data collection module determines the specific environmental monitoring area and collects multi-source heterogeneous data in the environmental monitoring area in real time.

[0042] Define the longitude and latitude boundaries of the enterprise factory and its surrounding sensitive areas, such as sewage outlets, residential areas, and ecological protection areas, delimit the environmental monitoring area, establish a three-dimensional monitoring network at the environmental monitoring area, divide the time points by t duration, and collect multi-source heterogeneous data of the environmental monitoring area at each time point, including establishing a ground sensing layer by deploying a temperature and humidity adaptive compensation sensor array, collecting the change data of PM2.5 and sulfur dioxide concentrations in each time period in the environmental monitoring area, dividing the monitoring area into A×A grids, deploying 1 sensor node at the center of each grid, and each sensor node contains a PM2.5 sensor and a sulfur dioxide sensor, and encrypt the boundary area to a spacing. In this embodiment, A is set to 100m and a is set to 50m. To avoid interference from temperature and humidity on sensor measurement, a temperature and humidity compensation formula is constructed, and the output value of the sensor is processed through the temperature and humidity compensation formula to improve the accuracy of data collection. The temperature and humidity compensation formula is specifically as follows:

[0043] S(x) = s(x) × [1 + a1 × (H(x) - h) + a2 × (R(x) - r)];

[0044] Where x represents the xth time point, S(x) represents the sensor output value after temperature and humidity compensation at the xth time point, s(x) represents the original sensor output value at the xth time point, H(x) represents the temperature value at the xth time point, h represents the standard temperature, R(x) represents the humidity value at the xth time point, r represents the standard humidity, a1 is the temperature compensation coefficient, representing the influence degree of temperature change on sensor output, and a2 is the humidity compensation coefficient, representing the influence degree of humidity change on sensor output.

[0045] Furthermore, the temperature compensation coefficient and the humidity compensation coefficient are obtained through experiments. By using a thermo-hygrostat, the temperature and humidity are precisely controlled. At each combination of temperature and humidity, the sensor output value is recorded. Meanwhile, a high-precision reference instrument (such as a NIST-certified thermo-hygrometer) is used to measure the true temperature and humidity values. The sensor output deviation is calculated. Through multiple linear regression, the influence coefficients a1 and a2 of temperature and humidity on the deviation are fitted. The compensation effect is verified on an independent test set. If the error exceeds the preset allowable range, the parameters are readjusted. In this embodiment, the allowable range is set to ±3%.

[0046] Edge computing technology is adopted to achieve real-time monitoring of PM2.5 and sulfur dioxide concentrations. By accessing the real-time data stream of NASAEarthdata, the Landsat-9 multispectral data of the environmental monitoring area is collected in real time. By parsing the Landsat-9 multispectral data, the normalized difference vegetation index (NDVI) is constructed using the red light band and the near-infrared band. The time period is divided by T duration. The NDVI of the same time period in different historical time periods is compared, and the vegetation cover change rate is calculated.

[0047] The change data of PM2.5, sulfur dioxide concentration, normalized difference vegetation index (NDVI), and vegetation cover change rate are collectively referred to as environmental change data.

[0048] The objective function setting module establishes the environmental monitoring management objective function for the environmental monitoring area and sets the constraint conditions for the environmental monitoring area based on the historical work intensity data of the enterprise factory.

[0049] Collect the environmental change data of each environmental monitoring area in N time periods. Use a unique number to identify each environmental monitoring area in the enterprise factory. Each number corresponds to only one environmental monitoring area, and the obtained set of serial numbers representing each area of the enterprise factory is denoted as {1, …, i, …, n}, where i represents the i-th environmental monitoring area and n represents that there are n environmental monitoring areas in the enterprise factory. On the premise of meeting the normal output of the enterprise factory, the environmental monitoring management objective function of the enterprise factory is established with the goal of minimizing the environmental resource consumption of each area of the enterprise factory. The environmental monitoring management objective function is specifically as follows:

[0050]

[0051] Among them, i represents the i-th area, so(i) represents the average sulfur dioxide concentration in the i-th area within a time period, Δpm(i) represents the number of time periods during which the PM2.5 concentration in the i-th area decreases within a time period, nv(i) represents the average value of the normalized difference vegetation index in the i-th area within a time period, Δfg(i) represents the number of time periods during which the normalized difference vegetation index in the i-th area increases within a time period, g(i) represents the pollution weight coefficient of the i-th area, and h(i) represents the greening weight coefficient of the i-th area.

[0052] Furthermore, the specific calculation formula of Δpm(i) is as follows:

[0053]

[0054] Among them, i represents the i-th area, x represents the x-th time point, N represents the total number of time points within a time period, pm(i,x) represents the PM2.5 concentration value collected at the x-th time point in area i, H(*) represents the step function, when * is less than or equal to 0, H(*) is equal to 0, and when * is greater than 0, H(*) is equal to 1.

[0055] Furthermore, the specific calculation formula of Δfg(i) is as follows:

[0056]

[0057] Among them, i represents the i-th area, x represents the x-th time point, N represents the total number of time points within a time period, nv(i,x) represents the normalized difference vegetation index collected at the x-th time point in area i, H(*) represents the step function, when * is less than or equal to 0, H(*) is equal to 0, and when * is greater than 0, H(*) is equal to 1.

[0058] According to the functional requirements of different enterprise factory areas, in order to ensure the equipment operation requirements, production capacity benchmark requirements, and energy consumption intensity requirements of each environmental area, different constraint conditions are set for different environmental monitoring areas;

[0059] Among them, the equipment operation requirements of area i are specifically as follows:

[0060]

[0061] Among them, m represents the m-th time period, M represents the total number of time periods included within a time period, zx(m) represents the equipment online rate in the m-th time period. The equipment online rate refers to the probability that the equipment remains online within a certain time, that is, the probability of normal operation and communication with the system. G is the online rate threshold, which is freely set according to the working requirements of the equipment in different environmental monitoring areas. I[*] is the conditional function, when the condition * is satisfied, I[*] is equal to 1, and A is the equipment operation requirement threshold.

[0062] The benchmark demand for production capacity in region i is specifically as follows:

[0063]

[0064] Among them, m represents the m-th time period, M represents the total number of time periods included in a time cycle, Q(i, m) represents the actual production capacity of region i in the m-th time period, with the unit of product quantity / output value (such as tons, pieces), Q(i) represents the benchmark value of the production capacity of the i-th region, that is, the normal production level in the same historical period or the production capacity stipulated in the contract. represents the production capacity weight coefficient of the i-th region, which is determined by the ratio of the total production capacity of this region in a time cycle to the production capacities of other regions of the enterprise. a is a constant, and in this embodiment, a is set to be equal to 1.

[0065] The energy consumption intensity demand for region i is specifically as follows:

[0066]

[0067] Among them, m represents the m-th time period, M represents the total number of time periods included in a time cycle, E(i, m) represents the total energy consumption of region i in the m-th time period. In this embodiment, electricity consumption is used to represent it. GD(i, m) represents the total economic output value of region i in the m-th time period, which is obtained from the enterprise's financial data. E(i) represents the benchmark energy consumption intensity, that is, the energy consumption standard per unit of GDP. For example, if the benchmark value of a certain region is 0.5 tons / 10,000 yuan, then for every 10,000 yuan of GDP created, at most 0.5 tons of standard coal are allowed to be consumed. ΔT represents the difference between the average temperature of the working equipment in region i and the benchmark temperature within a time cycle. The benchmark temperature is taken as the average temperature of the region in the same historical time cycle. β(i) represents the energy consumption weight coefficient of region i, and b is the temperature influence coefficient, representing the adjustment ratio of the benchmark value of the energy consumption intensity for every 1°C change in temperature. In this embodiment, b is set to be equal to 0.05.

[0068] The environmental prediction management module, based on the historical environmental change data of the enterprise factory's environmental monitoring area, establishes an environmental change data prediction model, and realizes the environmental monitoring and control of the enterprise factory by predicting the environmental change data.

[0069] Extract the historical environmental change data, and establish an environmental change data prediction model based on the historical environmental change data of consecutive historical time periods. The specific formula is as follows:

[0070] yc(i, m) = y0 + y1 × [y(i, m - 1)] 2 +…+ y k × [y(i, m - 1)] k ;

[0071] Among them, m represents the m-th time period, i represents the i-th region, y0, y1, y k represent model parameters, k represents the model order. By inputting historical environmental change data into the environmental change data prediction model, a predicted value is obtained. Based on the true value of the energy change data corresponding to the time period, the target squared error between the predicted value and the true value is determined. A target squared error threshold J is set. When the target squared error between the predicted value and the true value is less than or equal to the threshold J, it is determined that the training of the environmental change data prediction model is completed. When the target squared error between the predicted value and the true value is greater than the threshold J, it is determined that the training of the environmental change data prediction model is not completed. By adjusting the parameter values and the order of the model, the target squared error between the predicted value and the true value is reduced.

[0072] Using the trained prediction model, predicted values for the next W time periods are generated based on the environmental change data of the current time period. A pollution ratio threshold O is set, and the time periods during which the PM2.5 concentration or sulfur dioxide concentration rises in the predicted future time periods are determined and recorded as polluted time periods. If the ratio of the number of polluted time periods to W is greater than or equal to the threshold O, it is determined that the current pollution is severe and emission reduction measures or production plan adjustments are required.

[0073] As described above, only the preferred specific embodiments of the present invention are provided, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention, according to the technical solution and inventive concept of the present invention, makes equivalent substitutions or changes, and should be covered by the protection scope of the present invention.

[0074] The preferred embodiments of the present invention disclosed above are only used to help illustrate the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the present invention to only the specific embodiments. Obviously, many modifications and variations can be made according to the content of this specification. These embodiments are selected and specifically described in this specification to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and utilize the present invention well. The present invention is only limited by the claims and their full scope and equivalents.

Claims

1. An intelligent environmental monitoring and management system based on big data, characterized in that, Including: An environmental change data collection module, which is used to determine a specific environmental monitoring area and collect multi-source heterogeneous data of the environmental monitoring area in real time; A target function setting module, which is used to establish an environmental monitoring management target function for the environmental monitoring area and set the constraint conditions for the environmental monitoring area based on the historical work intensity data of the enterprise factory; An environmental prediction management module, which is based on the historical environmental change data of the environmental monitoring area of the enterprise factory, establishes an environmental change data prediction model, and realizes the environmental monitoring control of the enterprise factory by predicting the environmental change data.

2. The intelligent environment monitoring and management system based on big data according to claim 1, characterized in that, To determine a specific environmental monitoring area and collect multi-source heterogeneous data of the environmental monitoring area in real time, the specific method is as follows: Define the enterprise factory and its surrounding sensitive areas, delimit the environmental monitoring area, construct a three-dimensional monitoring network at the environmental monitoring area, divide the time points by t duration, and collect the multi-source heterogeneous data of the environmental monitoring area at each time point. The specific method is to establish a ground sensing layer by deploying a temperature and humidity adaptive compensation sensor array, collect the change data of PM2.5 and sulfur dioxide concentrations in each time period in the environmental monitoring area, divide the monitoring area into A×A grids, deploy 1 sensor node at the center of each grid, and each sensor node contains a PM2.5 sensor and a sulfur dioxide sensor. The boundary area is encrypted to a spacing, construct a temperature and humidity compensation formula, and process the output value of the sensor through the temperature and humidity compensation formula; By accessing the real-time data stream of NASA Earthdata, collect the Landsat-9 multi-spectral data of the environmental monitoring area in real time. By parsing the Landsat-9 multi-spectral data, construct a normalized difference vegetation index using the red light band and the near-infrared band, divide the time period by T duration, and compare the normalized difference vegetation indices in the same time period of different historical time periods.

3. The intelligent environment monitoring and management system based on big data according to claim 2, characterized in that, To construct a temperature and humidity compensation formula, the specific method is as follows: Use S(x) = s(x)×[1 + a1×(H(x) - h) + a2×(R(x) - r)] to represent the temperature and humidity compensation formula, where x represents the xth time point, S(x) represents the sensor output value after temperature and humidity compensation at the xth time point, s(x) represents the original sensor output value at the xth time point, H(x) represents the temperature value at the xth time point, h represents the standard temperature, R(x) represents the humidity value at the xth time point, r represents the standard humidity, a1 is the temperature compensation coefficient, and a2 is the humidity compensation coefficient.

4. The intelligent environmental monitoring and management system based on big data according to claim 1, characterized in that, To establish an environmental monitoring management target function for the environmental monitoring area, the specific method is as follows: Collect the environmental change data of each environmental monitoring area in N time periods, and use a unique number to identify each environmental monitoring area in the enterprise factory. Each number corresponds to only one environmental monitoring area, and the obtained set of serial numbers representing each area of the enterprise factory is denoted as {1, …, i, …, n}, where i represents the i-th environmental monitoring area and n represents that there are n environmental monitoring areas in the enterprise factory. Using the formula represents the environmental monitoring management objective function, where i represents the i-th area, so(i) represents the average sulfur dioxide concentration of the i-th area in a time period, Δpm(i) represents the number of time periods with a decrease in the PM2.5 concentration in the i-th area in a time period, nv(i) represents the average value of the normalized difference vegetation index of the i-th area in a time period, Δfg(i) represents the number of time periods with an increase in the normalized difference vegetation index of the i-th area in a time period, g(i) represents the pollution weight coefficient of the i-th area, and h(i) represents the greening weight coefficient of the i-th area.

5. The intelligent environmental monitoring and management system based on big data according to claim 4, characterized in that, The number of time periods when the PM2.5 concentration decreases, including: Using the formula represents the number of time periods during which the PM2.5 concentration decreases. Here, i represents the i-th region, x represents the x-th time point, N represents the total number of time points within a time period, pm(i, x) represents the PM2.5 concentration value collected at the x-th time point in region i, and H(*) represents the step function. When * is less than or equal to 0, H(*) equals 0; when * is greater than 0, H(*) equals 1.

6. The intelligent environment monitoring and management system based on big data according to claim 4, characterized in that The number of time periods when the normalized difference vegetation index increases, the specific method is as follows: Using the formula which represents the number of rising time periods of the normalized difference vegetation index, where i represents the i-th region, x represents the x-th time point, N represents the total number of time points within a time period, nv(i, x) represents the normalized difference vegetation index collected at the x-th time point in region i, and H(*) represents the step function, where H(*) equals 0 when * is less than or equal to 0 and H(*) equals 1 when * is greater than 0.

7. The intelligent environment monitoring and management system based on big data according to claim 1, characterized in that, To set the constraint conditions for the environmental monitoring area, the specific method is as follows: Using inequalities represent the device operation requirements for area i, where m represents the m-th time period, M represents the total number of time periods in a time cycle, zx(m) represents the device online rate in the m-th time period, G is the online rate threshold, I[*] is a conditional function that equals 1 when condition * is satisfied, and A is the device operation requirement threshold; Using inequalities represents the benchmark demand for production capacity in region i, where m represents the m-th time period, M represents the total number of time periods in a time cycle, Q(i, m) represents the actual production capacity of region i in the m-th time period, and Q(i) represents the benchmark value of production capacity for the i-th region. represents the production capacity weight coefficient for the i-th region, and a is a constant. Using inequalities represents the energy consumption intensity requirement of region i. E(i,m) represents the total energy consumption of region i in the m-th time period, GD(i,m) represents the total economic output value of region i in the m-th time period, E(i) represents the baseline energy consumption intensity, i.e., the energy consumption standard per unit of GDP, ΔT represents the difference between the average temperature of the working equipment in region i and the baseline temperature within a time cycle, β(i) represents the energy consumption weight coefficient of region i, and b is the temperature influence coefficient.

8. The intelligent environment monitoring and management system based on big data according to claim 1, characterized in that, To establish an environmental change data prediction model, the specific method is as follows: Extract historical environmental change data, establish an environmental change data prediction model based on the historical environmental change data of consecutive historical time periods, and use the formula yc(i,m) = y0 + y1×[y(i,m - 1)] 2 +…+y k ×y(i,m - 1)] k represents the environmental change data prediction model, where m represents the m-th time period, i represents the i-th region, y0, y1, y k represent the model parameters, k represents the model order. By inputting the historical environmental change data into the environmental change data prediction model, the predicted value is obtained. Based on the true value of the energy change data in the corresponding time period, the target squared error between the predicted value and the true value is determined. Set the target squared error threshold J. When the target squared error between the predicted value and the true value is less than or equal to the threshold J, it is judged that the training of the environmental change data prediction model is completed. When the target squared error between the predicted value and the true value is greater than the threshold J, it is judged that the training of the environmental change data prediction model is not completed. By adjusting the parameter values and order of the model, the target squared error between the predicted value and the true value is reduced.

9. The intelligent environmental monitoring and management system based on big data according to claim 1, characterized in that, To realize the environmental monitoring control of the enterprise factory by predicting the environmental change data, the specific method is as follows: Use the trained prediction model to generate predicted values for the next W time periods based on the environmental change data in the current time period. Set the pollution ratio threshold O, determine the time periods during which the PM2.5 concentration or sulfur dioxide concentration rises in the predicted future time period, and record them as polluted time periods. If the ratio of the number of polluted time periods to W is greater than or equal to the threshold O, it is judged that the current pollution is severe and emission reduction measures need to be taken or the production plan needs to be adjusted.

Images