KR102631124B1 - Air pollutant emission calculating system for area source - Google Patents

Abstract

A system for calculating air pollutant emissions from cotton pollutants is provided, which calculates air pollutant emissions inside the cotton pollutants from outside the cotton pollutants. The air pollutant emission calculation system of the cotton pollution source includes a plurality of support units arranged along the circumference of the cotton pollution source, a pollutant sensor arranged in at least one of the plurality of support units, and a pollutant sensor distributed in the plurality of support units. Multiple standard material generation modules that generate and spread standard materials that cannot be measured, multiple standard material sensors distributed in multiple support units, and multiple standard material sensors distributed in multiple support units to collect atmospheric flow information and environmental information. From the environmental sensors, atmospheric flow information, environmental information, and measured standard substance concentrations, the concentration change pattern of the standard substance that appears according to the change in position on the ground surrounded by the support unit is calculated, and the standard substance is calculated based on the measured pollutant concentration. It includes a calculation module that performs calculations to predict the concentration of pollutants at a specific point inside the cotton pollutant according to the concentration change pattern.

Description

Air pollutant emission calculation system for area source}

The present invention relates to a system for calculating the amount of air pollutants emitted from an air pollution source. More specifically, the present invention relates to a system for calculating the amount of pollutants emitted from an air pollution source. More specifically, the present invention relates to a system for calculating the amount of pollutants emitted from an air pollution source from the outside, without accessing the inside of the cotton source. It is about a pollutant emissions calculation system.

Sources that produce air pollutants are diversely distributed on the Earth's surface. In general, pollution sources can be classified into point sources, are sources, and line sources. Depending on the discharge type, they can also be classified into fixed and mobile sources.

Surface pollution sources may be fixed pollution sources located within a certain area and are known to have a direct impact on the spread of air pollutants on the ground surface. Examples of contamination sources vary, but may include, for example, livestock farms or some factories (which may be open without separate outlets). In livestock housing, ammonia can be the main emission.

These cotton pollutants are the source of air pollution and are therefore subject to continuous monitoring, but for various reasons (violation of private ownership guidelines, negative effects on livestock in livestock farms, etc.), it is also difficult to access the inside of the pollutants. As a result, situations often arise where internal pollutants must be measured outside the pollutant source.

However, differences in pollutant concentration between the inside and outside of cotton pollutants are inevitable, and there are limits to indirect predictions, so it is difficult to accurately calculate emissions using conventional general measuring devices (e.g., Korea Utility Model 20-0465003, etc.). there was. Therefore, technological alternatives are required.

Meanwhile, the inventors of the present invention received support from the National IT Industry Promotion Agency funded by the government (Ministry of Science and ICT) in 2021 to conduct research (No. S0837-21-1034, Development of an IoT monitoring system for reducing fine dust in small-scale emission workplaces) ) was performed.

Republic of Korea Registered Utility Model Publication No. 20-0465003, (2013. 02. 06), Specification

The technical problem of the present invention is to solve this problem, and is to provide a system for calculating the air pollutant emissions from cotton pollutants that can accurately calculate the pollutant emissions inside cotton pollutants from the outside without accessing the inside of the cotton pollutants.

The technical problem of the present invention is not limited to the problems mentioned above, and other technical problems not mentioned will be clearly understood by those skilled in the art from the description below.

The system for calculating air pollutant emissions from cotton pollutants according to the present invention calculates the air pollutant emissions from cotton pollutants for at least one point inside the cotton pollutants outside of the cotton pollutants occupying a certain area on the ground. In the emission calculation system, a plurality of support units arranged to be spaced apart from each other along the perimeter of the surface pollution source; a pollutant sensor disposed on at least one of the plurality of support units to measure pollutant concentration; A plurality of standard material generation modules distributed in the plurality of support units and generating standard materials that are not measured by the contaminant sensor and spreading them around the support units; A plurality of standard substance sensors distributed in the plurality of support units to measure the concentration of the standard substance; A plurality of environmental sensors distributed in the plurality of support units to collect atmospheric flow information and environmental information about the support units and their surroundings; And from the atmospheric flow information, the environmental information, and the measured concentration of the standard substance, a pattern of concentration change of the standard substance that appears according to the change in position on the ground surrounded by the support unit is calculated, and the measured concentration of the pollutant is used as the standard. It includes a calculation module that performs a calculation to predict the concentration of pollutants at a specific point inside the cotton pollutant according to the concentration change pattern of the standard substance.

The calculation module learns the correlation between the atmospheric flow information, the environmental information, and the measured concentration of the standard substance using an artificial intelligence learning algorithm to calculate the concentration change pattern of the standard substance, and the concentration change of the standard substance. The pattern may include a concentration change pattern that has a non-linear correlation with the change in location on the land surface.

The plurality of support units may be arranged irregularly around the surface contamination source.

The artificial intelligence learning algorithm includes an artificial neural network algorithm, and the artificial neural network algorithm includes at least one of a feed-forward neural network, a deep neural network, and a long short-term memory. It can contain one.

The contaminant sensor may be formed in plural, and at least one of the contaminant sensor, the standard substance generation module, the standard substance sensor, and the environmental sensor may be disposed in one of the support units.

The support unit includes a reference plate on which one of the standard material generation modules is located, and at least one standard material sensor formed at each vertex formed in a cubic shape capable of accommodating the standard material generation module and connected to the reference plate. It may include a reference material sensor support.

The standard material sensor support may have a structure that is repeated in at least one of the direction of gravity and the direction perpendicular to gravity with respect to the standard material generation module.

The environmental sensor may include a wind vane, anemometer, position sensor, and altimeter.

The environmental sensor may further include a thermometer, a hygrometer, and an optical sensor.

The contaminant may include at least one of ammonia and hydrogen sulfide, and the standard material may include at least one of carbon monoxide and oxygen.

According to the present invention, it is possible to more accurately calculate the amount of air pollutants generated inside the cotton pollutant from equipment placed outside the cotton pollutant. Therefore, it becomes possible to monitor air pollutant emissions from specific cotton pollutants that were previously difficult to access, and to manage and supervise the pollutants in the area. In particular, the present invention provides more accurate predicted values of emissions in various situations where prediction was difficult in the past through data analysis and learning using artificial intelligence, making it possible to calculate air pollutant emissions at a level different from that of the past.

Figure 1 is a plan view illustrating the arrangement of an air pollutant emission calculation system for cotton pollution sources according to an embodiment of the present invention.
Figure 2 is a side view showing an example of applying the air pollutant emissions calculation system of Figure 1 to a sloped ground surface.
Figure 3 is a conceptual diagram showing the calculation process of the concentration change pattern by the calculation module of the air pollutant emissions calculation system of Figure 1.
FIG. 4 is a diagram showing the structure of the support unit of the air pollutant emissions calculation system of FIG. 1 and its modification along with a pollutant sensor, a standard material sensor, and an environmental sensor.
Figure 5 is a diagram illustrating the measurement status of standard substances performed in each support unit.
Figures 6 and 7 are diagrams illustrating measurement situations of standard materials in which a plurality of support units are linked.

Advantages and features of the present invention and methods for achieving them will become clear by referring to the embodiments described in detail below along with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below and may be implemented in various different forms, and the present embodiments are only provided to ensure that the disclosure of the present invention is complete and to provide common knowledge in the technical field to which the present invention pertains. It is provided to fully inform those who have the scope of the invention, and the invention is merely defined by the claims. Like reference numerals refer to like elements throughout the specification.

In this specification, the 'air pollutant emissions calculation system from Myeon pollutants' may also be referred to as 'the air pollutant emissions calculation device from Myeon pollutants.'

Hereinafter, with reference to FIGS. 1 to 7, the system for calculating air pollutant emissions from cotton pollutants according to the present invention will be described in detail.

Figure 1 is a plan view illustrating the arrangement of an air pollutant emission calculation system for surface pollution sources according to an embodiment of the present invention, and Figure 2 shows an example of applying the air pollutant emission calculation system of Figure 1 to a sloping ground surface. This is a side view.

Referring to FIG. 1, the air pollutant emission calculation system 1 of a cotton pollutant according to the present invention includes a plurality of support units 10 arranged around the cotton pollutant source S. The support units 10 collect (through sensors) the information necessary to predict the emission of cotton pollutants (S), and the collected information is transmitted to the calculation module 30 and calculated.

The information collected from each support unit 10 includes information about standard substances that are similar to pollutants. The collected information is analyzed and calculated into a pattern that can predict the concentration of pollutants that fluctuate due to diffusion.

Patterns that can predict pollutant concentrations are calculated by learning information about standard substances, atmospheric flow information, and environmental information using an artificial intelligence learning algorithm. Therefore, meaningful patterns are generated even from data (concentration, location on the ground, wind direction, wind speed, altitude, temperature, humidity, solar radiation, etc.) for which the correlation is not clearly derived using conventional analysis methods (linear regression analysis, etc.) or have non-linear correlations. So it can be used for accurate predictions.

The air pollutant emission calculation system (1) of cotton pollutants according to the present invention is specifically configured as follows. The air pollutant emission calculation system (1) of the cotton pollution source calculates the air pollutant emissions from the outside of the cotton pollution source (S), which occupies a certain area on the ground, to at least one point inside the cotton pollution source (S). In the air pollutant emission calculation system (1), a plurality of support units (10) are disposed spaced apart from each other along the perimeter of the surface pollution source (S), and are disposed on at least one of the plurality of support units (10). A pollutant sensor (21) that measures the concentration of pollutants is dispersed in a plurality of support units (10), and generates standard substances that cannot be measured by the pollutant sensor (21) and spreads them around the support unit (10). A standard material generation module (22), a plurality of standard material sensors (23) distributed in a plurality of support units (10) to measure the concentration of the standard material, and a support unit (23) distributed in a plurality of support units (10). 10) and a plurality of environmental sensors 24 that collect surrounding air flow information and environmental information, and on the ground surrounded by a support unit 10 from the air flow information, the environmental information, and the measured concentration of the standard substance. Calculate the concentration change pattern of the standard substance that appears according to the change in location, and calculate the concentration of the pollutant at a specific point inside the cotton pollutant (S) according to the concentration change pattern of the standard substance based on the measured pollutant concentration. Includes module 30.

In one embodiment of the present invention, the calculation module 30 learns the correlation between atmospheric flow information, environmental information, and measured concentration of standard substances using an artificial intelligence learning algorithm to calculate the concentration change pattern of the standard substance. The concentration change pattern may include a concentration change pattern that has a non-linear correlation with the change in location on the ground.

In addition, in one embodiment of the present invention, the pollutant sensor 21 is formed in plural and includes a pollutant sensor 21, a standard material generation module 22, and a standard material sensor 23 in one support unit 10. , and environmental sensors 24 may be arranged at least one by one. Hereinafter, the configuration and effects of the present invention will be described in more detail based on an embodiment of the present invention.

Cotton pollution source (S) of the present invention refers to air pollution sources that appear as cotton on the ground. The surface pollution source (S) occupies a certain area on the ground and may be, for example, a factory or livestock farm. In this embodiment, a livestock farm is used as an example.

Since the surface pollution source (S) is formed on the surface, the elevation of the surface is reflected. That is, as shown in FIG. 2, in many cases, such as livestock farms formed in mountainous areas, the cotton contamination source (S) may not be in a completely flat shape. Therefore, when predicting the concentration of pollutants, it is necessary to consider diffusion in the direction perpendicular to the ground and the effects of changes in humidity and temperature due to elevation differences.

Since these precise conditions for the cotton pollutant source (S) were not previously considered, errors were severe even when internal emissions were predicted from the outside. In addition, the prediction method was only a method of interpolating or extrapolating external measurements based on simple distance proportions, so the errors were even more severe. The present invention converts all of these conditions into data, stores them, learns them (machine learning), and predicts material diffusion and concentration changes according to the learning results, so it is possible to very accurately calculate the air pollutant emissions from the cotton pollutant (S).

As shown in FIGS. 1 and 2, a plurality of support units 10 are arranged spaced apart from each other along the perimeter of the surface contamination source S. Each support unit (10) itself forms a measurement space that can determine the diffusion behavior of the material, and since all of the plurality of support units (10) are distributed and interconnected around the surface contamination source (S), the support units (10) ), a more expanded measurement space is formed at the same time. Therefore, it is possible to effectively collect information necessary for predicting emissions from individual support units 10 and/or multiple support units 10.

The pollutant sensor 21 may be placed in at least one of the plurality of support units 10. In this embodiment, the description is based on an example in which the contaminant sensor 21 is disposed in each support unit 10, but in other embodiments, all of the plurality of support units 10 are connected to each other to form an expanded measurement space. Considering the structure, it is also possible to place the pollutant sensor 21 only on the necessary support unit 10.

The pollutant sensor 21 is formed in various ways to measure target pollutants. For example, if the pollutant source (S) is a livestock farm, the main pollutants may be ammonia (NH ₃ ), hydrogen sulfide (H ₂ S), etc., and the pollutant sensor 21 is formed as a sensor capable of measuring the pollutants. It can be. Since the pollutant sensor 21 is not excluded from being a combination of multiple sensors, it may be formed by a combination of multiple sensors.

A plurality of standard substance generation modules (22) are distributed and arranged in a plurality of support units (10), and generate standard substances that cannot be measured by the contaminant sensor (21) and spread them around the support unit (10). In this embodiment, the standard material generation module 22 is also described based on an example in which each support unit 10 is disposed, but in other embodiments, the entire plurality of support units 10 are connected to each other to form an expanded measurement space. Considering the structure, it is also possible to arrange the standard material generation module 22 in a plurality of selected support units 10.

Since the standard material generation module 22 can be formed in various forms capable of generating standard materials, there is no need to be limited in structure or operation method. For example, the standard material can be taken in various possible forms depending on the standard material, such as a form containing a solid material that spreads the standard material particles by vaporization, a form containing the standard material gas in a container, and a form that generates the standard material by vaporizing it from a liquid. The generation module 22 can be implemented.

The standard substance is a substance that can be measured by a standard substance sensor, and preferably a substance that is not detected by the pollutant sensor 21 is used. Therefore, the pollutants can spread to the surrounding space without causing unnecessary operation of the pollutant sensor 21. In addition, it is desirable that the standard substance poses little risk to the human body or the environment. In addition, it is preferable that the standard material has a molecular weight similar to that of the target contaminant, and therefore different types of standard materials may be applied depending on the contaminant. For example, if the contaminant includes at least one of ammonia and hydrogen sulfide, the standard material may include at least one of carbon monoxide (CO) and oxygen (O ₂ ). Carbon monoxide is a toxic substance, but it can be used as a standard substance without any real damage if diluted to an appropriate concentration and dispersed in a large outdoor space. In other words, regardless of the hazard of the substance itself, if the hazard can be practically avoided, it can be applied as a standard substance.

If it is difficult to find a satisfactory standard material, it is possible to select an arbitrary material as a standard material and make additional corrections after determining the similarity in behavior between the contaminant and the selected standard material. For example, under the same conditions, pollutants and standard materials can be diffused into the same space, the difference in behavior between the two can be identified using sensors for each, and additional corrections can be made by calculating a correction coefficient that corrects the difference. . During such corrections, it is also possible to learn the data using an artificial intelligence learning algorithm to derive the differences between them as patterns and perform corrections to remove the differences.

A plurality of standard substance sensors 23 are distributed and arranged in a plurality of support units 10 to measure the standard substance concentration. The standard substance sensor 23 does not need to be particularly limited in its method as long as it can measure standard substances. In particular, a plurality of standard material sensors 23 can be placed on one support unit 10, and as described later, they can be periodically arranged in space to form a structural arrangement surrounding the standard material generation module 22. there is. The structural features of the support unit 10 will be described in more detail later.

A plurality of environmental sensors 24 are distributed and arranged in a plurality of support units 10 to collect atmospheric flow information and environmental information around the support units 10. The environmental sensor 24 may also be formed as a combination of multiple sensors, for example, as a combination of the following devices.

The environmental sensor 24 may include a wind vane, anemometer, position sensor, and altimeter, and may further include a thermometer, hygrometer, and optical sensor as needed. That is, the environmental sensor 24 can be formed by a combination of various devices that can measure desired environmental conditions. However, there is no need to be limited as such, and the environmental sensor 24 can be implemented in a form where necessary environmental information is provided by being connected to a national computer network, etc. through data communication. However, in this embodiment, the description is made in the form of a combination of multiple devices.

The environmental sensor 24 can collect atmospheric flow information in the measurement area using a wind vane and wind speed gauge on the support unit 10 arranged around the surface pollution source (S), and collects information about the location and elevation on the ground using a position measuring device ( e.g. GPS devices) and altimeters. Additionally, information on temperature, humidity, and solar radiation (measured from light intensity) can be collected using thermometers, hygrometers, and optical sensors.

Therefore, environmental information can include most of the obtainable information about the area where the pollutant source (S) is located and its surrounding environment, such as location and elevation on the ground, temperature, humidity, and solar radiation. In addition to what is listed, environmental information can be added as needed using corresponding measuring devices, etc.

The calculation module 30 determines the atmospheric flow information and environmental information collected from the environmental sensor 24, and the standard substance concentration measured by the standard substance sensor 23, according to the change in position on the ground surrounded by the support unit 10. Calculate the concentration change pattern of the standard substance that appears. In addition, the calculation module 30 is a calculation diagram that predicts the concentration of pollutants at a specific point inside the cotton pollutant source (S) according to the concentration change pattern of the calculated standard substance based on the pollutant concentration measured by the pollutant sensor 21. run in parallel The calculation process and configuration of the calculation module 30 will be described in more detail with reference to FIG. 3.

Figure 3 is a conceptual diagram showing the calculation process of the concentration change pattern by the calculation module of the air pollutant emissions calculation system of Figure 1.

First, the calculation module 30 may be formed as a type of computer device. Calculations performed in the calculation module 30 may be performed by one or more calculation programs executed on one or more CPUs. The calculation module 30 can be placed remotely away from the support unit 10, but is connected to the contaminant sensor 21, standard material sensor 23, and environmental sensor 24 located in the support unit 10 by wired or wireless connection. can be connected to enable data communication. Therefore, as shown in FIG. 3, the data collected and transmitted by each of the pollutant sensor 21, the standard material sensor 23, and the environmental sensor 24 can be calculated using an algorithm.

In particular, the calculation module 30 repeatedly learns data transmitted from each sensor using an artificial intelligence learning algorithm. It is also possible to understand the relationship between multiple data as a kind of pattern and classify the pattern through pattern learning and comparison of learned patterns. Through machine learning using this artificial intelligence learning algorithm, the calculation module 30 is a standard that appears according to the change in position on the surface (i.e., the measurement space where the surface contamination source is located) surrounded by the support unit (see 10 in Figures 1 and 2). Calculate the concentration change pattern of the substance.

The artificial intelligence learning algorithm of the calculation module 30 includes an artificial neural network (ANN) algorithm in which at least one hidden layer consisting of a plurality of calculation nodes is formed between the input layer and the output layer, and the artificial neural network algorithm is a forward It may include at least one of a feed-forward neural network, a deep neural network, and a long short-term memory.

To explain in more detail, in an artificial neural network algorithm, the nodes of each layer may be weighted functions. These constitute artificial neurons in the form of a network, and calculations are performed through their synthesis according to an algorithm. Data is input to the nodes of the input layer, passes through a plurality of nodes formed in the hidden layer, and is propagated to the nodes of the output layer for calculation.

During the calculation process, the weight of the node can be updated by back propagation, etc. to optimize the result. Therefore, from the results of calculations, for example, modeling that shows various correlations between input data, such as non-linear correlation, is possible, and correlations and similarities between data that are difficult to easily identify in the past from multiple and various types of input data are also identified. By appropriately adjusting or selecting each node or function that synthesizes the nodes or processes the results of the node, an artificial neural network that performs the desired type of operation can be constructed.

Forward neural network, deep neural network, and long-short-term memory are among the known artificial neural networks, and each has the following characteristics. For example, forward neural networks can be used appropriately for approximation and estimation, and deep neural networks can be effective in extracting more core content from complex data through deep learning. Long-short-term memory is an improvement on the recurrent neural network (RNN), and is advantageous for processing data that includes time like a recurrent neural network. Since it considers long-term data rather than a recurrent neural network, it can be effective in path prediction, etc.

The calculation module 30 learns the data collected around the cotton pollutant source using at least one of these artificial neural network algorithms to calculate the concentration change pattern of the standard substance that appears according to the change in location on the ground around and/or inside the cotton pollutant source. . With the neural networks described above, learning can also include predictions based on collected data. The concentration change pattern may be composed of a combination of learned data, and reflects the totality of information about the measurement space collected around the pollutant source, so fluctuating information (e.g., atmospheric flow information such as wind direction and speed, temperature, humidity, etc. , solar radiation, etc.) can be calculated in multiple ways in response to a set of environmental information.

For example, under environmental conditions (environmental condition 1) where temperature, humidity, solar radiation, wind direction, and wind speed are specified, a concentration change pattern (concentration change pattern 1) showing the concentration change distribution of the standard substance according to the change in location on the ground is calculated. Then, in environmental conditions (environmental condition 2) where temperature, humidity, solar radiation, wind direction, and wind speed have changed, a changed concentration change pattern (concentration change pattern 2) in which the concentration change distribution of the standard substance has changed according to the change in location on the ground is calculated. It can be. These calculations can be repeated and updated during the data learning process. By calculating, storing, and updating countless concentration change patterns according to combinations of environmental conditions, pollutant emissions can be predicted with accurate concentration change patterns that match the environmental conditions at the time of pollutant discharge prediction.

The concentration change pattern of a standard substance is, for example, a kind of map that shows the distribution of the concentration change amount or change rate that changes due to diffusion when the location on the ground changes when environmental conditions such as temperature, humidity, solar radiation, wind direction, and wind speed are specified. It may have appeared as . (As mentioned above, learning involves prediction, so the distribution of concentration change or change rate may at least partially include data predicted during the learning process.) For example, one point in the measurement space (the space on the ground where the surface pollution source is located). It is also possible to calculate a concentration change pattern in which the ratio of concentration change corresponding to the displacement between different points is defined for each point.

Therefore, if you trace the change in concentration in space according to the concentration change pattern, you can conveniently predict the emission of pollutants at a specific point inside the pollutant source from the pollutant concentration value measured outside the pollutant source.

In other words, the emission of pollutants at a specific point inside the cotton pollution source can be predicted according to the concentration change pattern of the standard substance calculated based on the pollutant concentration measured outside the cotton pollution source by the pollutant sensor 21. If the molecular weight of the standard substance and the pollutant are almost similar, the calculated concentration change pattern of the standard substance can be applied to the pollutant to predict the pollutant emissions at a point within the cotton pollutant.

In addition, as described above, if there is a difference between the pollutant and the standard material and it is difficult to apply it right away, correction to remove the difference is performed using the artificial intelligence learning algorithm, etc., and the calculated concentration change pattern is used to determine the pollutant source. I can predict my pollutant emissions.

The above concentration change pattern calculation and application method is a conceptual illustration of a specific situation, and the specific application method may vary depending on how the support unit is used. As mentioned above, each support unit itself forms a space where the diffusion behavior of the material can be identified, so it is also possible to calculate the concentration change pattern by learning the data collected from each support unit for each support unit and then combining the learning results of multiple support units. In addition, since all of the multiple support units distributed around the surface contamination source can be seen as forming an expanded measurement space (including the surface contamination source), data collected from all of the multiple support units can be learned at once. It is also possible to immediately calculate the concentration change pattern of the standard substance for the entire space containing the pollutant.

Below, the structure of the support unit will be described in more detail, and then the calculation and application method of the concentration change pattern of the standard substance will be explained in more detail.

FIG. 4 is a diagram showing the structure of the support unit of the air pollutant emissions calculation system of FIG. 1 and its modification along with a pollutant sensor, a standard material sensor, and an environmental sensor.

Referring to FIG. 4, the support unit 10 includes a reference plate 110 on which one standard material generation module 22 is located and a standard material sensor support 120 of a cubic shape capable of accommodating the standard material generation module 22. Includes. The standard material sensor support 120 is connected to the reference plate 110, and a standard material sensor 23 is placed at each vertex of the cube. The support unit 10 may include at least one reference material sensor support 120.

All reference material sensor supports 120 are not necessarily directly connected to the reference plate 110, but at least one is directly connected to the reference plate 110. The added standard material sensor support 120 may be coupled to another standard material sensor support 120 and indirectly connected to the reference plate 110. The connection between the reference material sensor support 120 and the reference plate 110 includes such direct and indirect connections.

One side of the standard material sensor support 120 may have a length in meters, for example. Therefore, each of the shapes illustrated in Figures 4 (a), (b), and (c) has a three-dimensional occupied space several meters long in the vertical and horizontal directions. Since this space is surrounded by a plurality of standard substance sensors 23 arranged at each vertex of the standard substance sensor support 120, it is very advantageous to measure the change in concentration of the standard substance diffused from the standard substance generation module 22.

The standard material sensor support 120 may have a structure that is repeated in at least one of the direction of gravity and the direction perpendicular to gravity with respect to the standard material generation module 22. Accordingly, for example, the support unit 10-1 is a multi-layered structure extending in the vertical direction as shown in (b) of FIG. 4, or a complex multi-layered structure extending in both vertical and horizontal directions as shown in (C) of FIG. 4. It can be modified in various ways, such as with a support unit (10-2).

Since the support unit 10 of this structure has a structure in which a plurality of standard material sensors 23 surround the standard material generation module 22 on the top, bottom, left, and right, the standard material sensor 23 located in the diffusion path of the standard material You can effectively measure the change in concentration of a standard substance due to a change in location. By repeatedly connecting multiple standard material sensor supports 120 or increasing the length of the side, the space occupied by the support unit 10 can be expanded, and the accuracy of the concentration change pattern of the standard material (since the actual measured space increases) can also be improved.

The pollutant sensor 21 and environmental sensor 24 can be placed at other points of the support unit 10, and additional devices 25 such as another sensor, data communication device, or power device can be conveniently placed in the remaining space. It can be placed. The support unit 10 of the present invention is itself configured to occupy space three-dimensionally.

Figure 5 is a diagram illustrating the measurement status of standard substances performed in each support unit.

Using the structure of the support unit 10, first, measure how the standard material generated by the standard material generation module 22 spreads and changes in concentration within the space occupied by the support unit 10 as shown in FIG. 5. can do. Since the standard material diffuses more in the wind (W) direction that changes on the plane as shown in (a) of FIG. 5, the standard material sensor 23 is roughly connected to the standard material generation module 22 (by the cubic standard material support). Even if they are spaced equally apart, different concentration values (A, B, C, D) can be detected.

Likewise, the standard material diffuses more in the direction of the wind (W), which changes on the side, as shown in Figure 5 (b). Therefore, the standard substance sensors 23 can detect different concentration values (A, B, A', B', A", B") even if they are separated from the standard substance generation module 22 at approximately equal intervals in the vertical direction. there is.

In this way, the standard substance sensors 23 can detect concentration changes three-dimensionally around the standard substance generation module 22. The measured concentration values are calculated from the calculation module (FIGS. 1 to 3) along with atmospheric flow information (wind direction and wind speed) and environmental information (location on the ground, altitude, temperature, humidity, solar radiation, etc.) collected by the environmental sensor 24 during measurement. (see 30) is transmitted and learned. Through repeated data collection, transmission, and learning processes, the calculation module 30 determines not only the points where each standard material sensor 23 is placed, but also points that can be predicted through learning, such as points between the standard material sensors 23. Concentration changes can also be calculated. Through this, the concentration change distribution of the standard substance can be calculated.

The concentration change distribution of these standard substances can be individually calculated using information collected from each of a plurality of different support units 10 and then synthesized together. Since all the support units (10) are adjacent to the perimeter of the surface pollution source, the concentration change distribution of the standard substance calculated for the support units (10) is the standard that appears according to the change in position on the ground surrounded by the support units (10). It can be converted and applied to the concentration change pattern of the substance. In this case, for example, the distance unit of the support unit 10 can be converted into an enlarged distance unit inside the cotton pollutant source through distance proportion and used to predict the concentration of pollutants.

Figures 6 and 7 are diagrams illustrating measurement situations of standard materials in which a plurality of support units are linked.

Meanwhile, the concentration change pattern may be calculated by utilizing all of the plurality of support units 10 distributed around the surface contamination source S, for example, as shown in FIG. 6.

Since the plurality of support units 10 are spaced apart from each other in the expanded space, the plurality of support units support the atmospheric flow (which can be thought of as a streamline continuously crossing the surface pollution source along the shown wind (W) direction) formed more widely in the space. It can be predicted from the units 10. For example, if each support unit 10 is viewed as a point in the entire space surrounding the surface contamination source S, and assuming a general situation where the wind direction does not change drastically when passing through adjacent points, data learning can be performed. Through this, it is also possible to predict the atmospheric flow passing through the surface pollution source (S). As described above, learning includes prediction, so the flow direction inside the unmeasured surface pollution source (S) can also be predicted through repeated learning of wind direction and wind speed data.

Since the atmospheric flow passing through the surface pollution source is predictable in this way, if the changes in the concentration values of the standard substance sensors 23 transmitted from each support unit 10 are learned at once, the surface pollution source surrounded by a plurality of support units 10 (S) and its entire surroundings can be learned as a single measurement space, so that the concentration change distribution of standard substances caused by atmospheric flow across the surface pollution source (S) can be directly calculated for the entire space.

That is, all of the plurality of support units 10 measure the concentration values of the standard material sensors 23 that are spread more in each direction by different winds (W), and the concentration measurements are based on atmospheric flow information (wind direction and Wind speed) and environmental information (location on the ground, altitude, temperature, humidity, location solar radiation, etc.) can be transmitted to the calculation module (see 30 in FIGS. 1 to 3) and learned all at once.

(The concentration measurements of standard substances that appear different due to differences in diffusion in each support unit are indicated as A1~D1, A2~D2, and A3~D3, respectively, and are the same as Figure 7.)

This concentration change pattern calculation method is not substantially different from the learning method for the space occupied by the support unit 10 described above. It can be seen that the space occupied by the support unit is the entire space where the cotton pollutants are distributed, and the standard material sensor surrounding the space occupied by the support unit has been changed to correspond to a plurality of support units surrounding the cotton pollutants.

Therefore, through data learning, it is possible to predict and calculate the concentration change distribution of standard substances in areas where the support units 10 are not placed (including the area inside cotton pollutants), and the concentration change of standard substances calculated for the entire extended space. The distribution can also be directly applied to the concentration change pattern of the standard substance that appears according to the change in location on the ground surrounded by the support unit 10.

If the concentration change pattern of the standard substance is calculated in this way, the pollutant concentration at a specific point inside the cotton pollutant (S) can be directly predicted by backcalculating from the pollutant concentration measured around the cotton pollutant (S).

Since the positions and relative distances of the support units 10 on the ground are accurately known by the position measuring device (GPS, etc.) included in the environmental sensor 24, there is no need to arrange them at exactly equal intervals. Therefore, as shown in FIG. 7, the support units 10 can be freely placed and measured at desired positions around the surface contamination source S-1 of an irregular shape.

Not only can the plurality of support units 10 be arranged irregularly around the surface contamination source S, but also individually collect data as described above to calculate the concentration change distribution for each occupied space. Since it is possible, it can be used by placing it individually. In other words, it is possible to measure by adding the number of support units 10 in the direction where diffusion occurs more, taking into account the wind (W) direction of the measurement space.

In this way, by arranging a plurality of support units 10 around various types of surface contamination sources (S), the concentration of contaminants at a specific point inside the contamination source (S) can be calculated.

Although embodiments of the present invention have been described above with reference to the attached drawings, those skilled in the art will understand that the present invention can be implemented in other specific forms without changing the technical idea or essential features. You will be able to understand it. Therefore, the embodiments described above should be understood in all respects as illustrative and not restrictive.

1: Calculation system for air pollutant emissions from pollutants in Myeon
10: Support unit 21: Pollutant sensor
22: Standard material generation module 23: Standard material sensor
24: Environmental sensor 25: Additional device
30: calculation module 110: reference plate
120: Standard material sensor support
S: Cotton Pollution source W: Wind

Claims (10)

In the air pollutant emission calculation system for cotton pollutants that calculates the air pollutant emissions for at least one point inside the cotton pollution source outside of the cotton pollution source occupying a certain area on the ground,
A plurality of support units arranged to be spaced apart from each other along the perimeter of the surface contamination source;
a pollutant sensor disposed on at least one of the plurality of support units to measure pollutant concentration;
A plurality of standard material generation modules distributed in the plurality of support units and generating standard materials that are not measured by the contaminant sensor and spreading them around the support units;
A plurality of standard substance sensors distributed in the plurality of support units to measure the concentration of the standard substance;
A plurality of environmental sensors distributed in the plurality of support units to collect atmospheric flow information and environmental information about the support units and their surroundings; and
From the atmospheric flow information, the environmental information, and the measured concentration of the standard substance, a concentration change pattern of the standard substance that appears according to the change in position on the ground surrounded by the support unit is calculated,
An air pollutant emission calculation system of a cotton pollutant source including an operation module that performs a calculation to predict the pollutant concentration at a specific point inside the cotton pollutant source according to the concentration change pattern of the standard substance based on the measured pollutant concentration. According to paragraph 1,
The calculation module learns the correlation between the atmospheric flow information, the environmental information, and the measured concentration of the standard substance using an artificial intelligence learning algorithm to calculate the concentration change pattern of the standard substance, and the concentration change of the standard substance. The pattern is a system for calculating air pollutant emissions from cotton pollutants, including a concentration change pattern with a non-linear correlation with the change in location on the land surface. According to paragraph 2,
A system for calculating air pollutant emissions from a cotton pollutant source, wherein the plurality of support units are irregularly arranged around the cotton pollutant source. According to paragraph 2,
The artificial intelligence learning algorithm includes an artificial neural network algorithm, and the artificial neural network algorithm includes at least one of a feed-forward neural network, a deep neural network, and a long short-term memory. A system for calculating air pollutant emissions from cotton pollutants, including one. According to paragraph 1,
The pollutant sensor is formed in plural, and at least one pollutant sensor, the standard material generation module, the standard material sensor, and the environmental sensor are disposed in one of the support units. An air pollutant emission calculation system of a pollutant source. . According to clause 5,
The support unit includes a reference plate on which one of the standard material generation modules is located, and at least one standard material sensor formed at each vertex formed in a cubic shape capable of accommodating the standard material generation module and connected to the reference plate. Air pollutant emission calculation system from cotton pollutants including standard material sensor support. According to clause 6,
The standard material sensor support is a surface pollutant emission calculation system that has a structure that is repeated in at least one of the direction of gravity and the direction perpendicular to gravity with respect to the standard material generation module. According to clause 5,
The environmental sensor is a system for calculating air pollutant emissions from cotton pollutants, including a wind vane, an anemometer, a position measurer, and an altimeter. According to clause 8,
The environmental sensor is a system for calculating air pollutant emissions from cotton pollutants, further comprising a thermometer, a hygrometer, and an optical sensor. According to paragraph 1,
The pollutant includes at least one of ammonia and hydrogen sulfide, and the standard material includes at least one of carbon monoxide and oxygen. A system for calculating air pollutant emissions from cotton pollutants.

Images