KR102393536B1 - Apparatus having auto correction algorism for measuring air quality and system therefor - Google Patents

Abstract

The present invention provides a light scattering measuring unit for measuring a first current based on scattered light scattered by emitting light to air in which suspended particles are filtered, and an alpha ionizing measuring unit for measuring a second ionized current by emitting alpha rays into the air. , Based on the first to third proportional constant set values set by calculating the concentration of air, fine dust and moisture in the beta-ray ionization measurement unit and the reference chamber for measuring the ionized third current by emitting beta-rays to the air, the The first concentration of the fine dust and the second concentration of moisture are determined for the first to third currents, and the final concentration of the fine dust is calculated by machine learning/deep learning calculation of the measurement error of the first and second concentrations. It provides an air quality measuring device including a calibration unit for determining.

Description

Air quality measuring device and system with built-in automatic calibration algorithm

The present invention relates to an air quality measuring device and system having an automatic calibration algorithm built-in, and more particularly, to an air quality measuring device and system for easily measuring the amount of fine dust contained in the air.

The technology to remove the noise of the sensor can be applied only when there is reference data that can determine whether the value of the sensor is correct. Also, it is mostly used to increase the accuracy of the sensor rather than to remove the noise generated in real time. In addition, the existing noise reduction technology cannot be applied in real time, but it must be carried out in advance at the stage of testing the sensor (or the stage of testing before the product using the sensor is shipped).

Existing technology can produce more accurate results the more data is accumulated, but it is not necessary to have a lot of data, but only when there is data under various conditions and various environments as much as possible for efficient noise removal and data correction.

However, reference data does not always exist in all environments in which sensors are tested and used, and it is difficult to predict all variable situations that will occur in the real environment. And it is difficult to obtain good results in terms of noise removal because it is possible to respond only to the variable conditions that occurred during the test period, not in real time.

For the above reasons, the existing noise cancellation technology is not focused on noise removal but rather on sensor calibration, so it is difficult to use it for actual noise cancellation.

In addition, particulate matter (PM) in the atmosphere is fine particles having various kinds of chemical properties. In particular, most of the fine dust, which directly or indirectly has a detrimental effect on the human body, is generated by artificial industrial activities.

Fine dust is fine dust that is invisible to the naked eye. It refers to dust with a diameter of 10㎛ or less. PM (Particulate Matter) means 'particulate matter (fine particles in a solid or liquid state floating in the atmosphere)'. Dust with a diameter of 1.0 μm or less. In the case of ultrafine dust, interest is growing because it can better penetrate into the alveoli and blood vessels, which can have a greater impact on health.

According to the air pollution process test standards, the beta-ray absorption method and the weight concentration method are specified as standard methods for measuring fine dust in the air environment.

The beta-ray absorption method is a method of measuring the mass concentration of fine dust by measuring the relative intensity of the beta-rays absorbed when the beta-rays emitted from the beta-ray source pass through the dust collected on the air filter. Due to the limitations of the measurement method of the indirect method, in order to derive an accurate mass value, it has the disadvantage of having to go through the gravimetric method and the comparative verification step.

The gravimetric concentration method is one of the methods for measuring the mass concentration of PM10 or PM2.5 in the atmospheric environment. Collect a sample of fine dust in the air using a sampler, and calculate the difference in weight between the filter before and after collection as the mass concentration. The gravimetric concentration method has limitations in that it is inconvenient to collect samples for a long time and cannot detect changes in concentration during collection.

Although it is not the standard method mentioned above, a light scattering method that can be read directly without taking a sample is also commonly used.

The light scattering method uses the principle that when light is irradiated to particulate matter (PM) suspended in the atmosphere, it is scattered by the particles. It is a technique for measuring the amount of particulate matter.

Among the methods using light scattering, the nephelometer method, which measures the amount of scattered particles using a light source having three wavelengths, and the spectrometer method, which measures the amount of scattered particles using a light source close to a single wavelength generated by a single wavelength such as a laser or an LED There is this.

The light scattering method has a limitation in that the accuracy of the light scattering method is reduced due to changes in the external environment in converting the light scattering degree into a number and using the calculated weight value due to fine dust. In particular, it has the greatest influence among external environmental factors that reduce the accuracy of moisture.

Recently, research is underway to measure the exact concentration of fine dust by increasing the accuracy of the moisture contained in the air.

KR 10-2018-0041828 A

An object of the present invention is to provide an air quality measuring device and system having an automatic calibration algorithm that is easy to measure the amount of fine dust contained in the air.

Another object of the present invention is to provide an air quality measuring device and system in which a plurality of sensors are used and an automatic calibration algorithm for selecting a sensor having an error such as noise in a measured value is built-in.

In addition, it is an object of the present invention to measure the amount of moisture and fine dust contained in the air by applying the light scattering method and the radiation method, and to correct the error according to the deep learning method, thereby making it easy to accurately measure the amount of fine dust It is to provide an air quality measuring device and system.

The objects of the present invention are not limited to the above-mentioned objects, and other objects and advantages of the present invention not mentioned may be understood by the following description, and will be more clearly understood by the examples of the present invention. It will also be readily apparent that the objects and advantages of the present invention may be realized by the means and combinations thereof indicated in the appended claims.

An air quality measurement system with an automatic calibration algorithm built-in according to an embodiment of the present invention includes: a sensor device having first to N-th sensor units of the same type for measuring the same object; and converting the first to N-th measured values of the first to N-th sensor unit into first to N-th positions having first to N-th distances in a polar coordinate system, and N using the first to N-th positions as vertices An error removing unit that calculates a discriminant value of a noise discriminant based on the length of each side of the square and determines whether all of the first to Nth sensor units are normal based on whether the discriminant value falls within a preset range and an air quality measuring device provided with, wherein the azimuth angles of the first to Nth positions may be integer multiples of a specific angle.

In addition, the discriminant is a sum of differences in lengths of the i-jth sides of the N-gon, wherein i is a natural number from 1 to N, and j is a natural number from any one of i to N. .

Also, the discriminant may be based on first to third measured values of the first to third sensor units.

In addition, the first sensor unit may include: a light scattering measuring unit measuring a first current based on the scattered light scattered by emitting light to the air in which the airborne particles are filtered in the measuring area; an alpha ionization measuring unit measuring a second current ionized by emitting alpha rays to the air; and a beta-ray ionization measuring unit measuring the ionized third current by emitting beta-rays to the air.

In addition, the air quality measuring device calculates the concentrations of air, fine dust, and moisture in the reference chamber, and based on first to third set values of proportional constants, the first to third currents of the fine dust Determining a second concentration of concentration and moisture, and deep learning calculation of the error of the first and second concentrations to determine a measurement error for the concentration of fine dust, and the first and second specific concentrations among the concentration data A reference concentration value in which data is extracted, and a first and second concentration difference value between the first and second specific concentrations of the fine dust included in each of the first and second specific concentration data and the concentration of the fine dust is set If it does not fall within the range, a calibration unit for determining the final concentration of the fine dust based on the difference between the first and second specific concentrations and the measurement error may be further provided.

The storage unit may further include a storage unit configured to store the first to third currents and the first to third proportional constant set values.

In addition, the calibration unit, based on the first to third currents and the first to third proportional constant set values, a concentration determination unit for determining the first and second concentrations for the first to third currents ; and a concentration measurement unit measuring the final concentration of the fine dust by deep learning operation on the measurement error of the first and second concentrations, wherein the concentration determining unit calculates the first to third currents using the following multiple linear regression equation can be used to calculate the first and second concentrations.

(IL is the first current, Iα is the second current, Iβ is the third current, A is the concentration value of air, P is the first concentration of fine dust, W is the second concentration of moisture, and Ka1, Ka2, and Ka3 are each The first set of proportional constant values, Kp1, Kp2, and Kp3, indicating the magnitude of the current set in proportion to the air concentration when light, alpha rays, and beta rays are emitted 2 Proportional constant set values, Kw1, kw2, and Kw3 are the third set of proportional constants indicating the magnitude of the current set in proportion to the moisture concentration when light, alpha, and beta rays are emitted)

In addition, the concentration measuring unit, the first predicted concentration of the fine dust and the first predicted concentration of the fine dust with respect to a convolutional neural network (CNN) based on the previously input first previous concentrations of fine dust and the second previous concentrations of moisture CNN learning processing unit for outputting a second predicted concentration of moisture; The third predicted concentration of the fine dust and the fourth predicted concentration of the moisture at the current time by applying the first previous time concentrations of fine dust and the second previous time concentrations of moisture input before the current time point to the set tanh function LSTM learning processing unit to output; and calculating the measurement error by comparing the first to fourth predicted concentrations with the first and second concentrations, and correcting the second concentration by the measurement error, so that the first concentration is finally corrected. It may be provided with an error correction unit for determining the final concentration of.

An air quality measuring device according to an embodiment of the present invention is a polar coordinate system for the first to Nth measurement values of the first to Nth sensor units of the sensor device having the same type of first to Nth sensor units for measuring the same object. Converting the first to Nth positions having the first to Nth distances, calculating the discriminant value of the noise discriminant based on the length of each side of the N-gon having the first to Nth positions as a vertex, and the and an error removing unit for determining whether all of the first to Nth sensor units are normal based on whether the determination value falls within a preset range, and the azimuth angle of the first to Nth positions may be an integer multiple of a specific angle .

The air quality measuring apparatus and system having an automatic calibration algorithm built-in according to the present invention can easily determine whether there is an error sensor among a plurality of sensors, and as there are more sensors, it is possible to distinguish a sensor causing an error.

This device and system can measure the final concentration of fine dust whose measurement error has been corrected by machine learning/deep learning calculations on the first to third currents measured by emitting light, alpha rays and beta rays into the air. It has the advantage of being able to accurately measure the concentration of

In addition to the above-described effects, the specific effects of the present invention will be described together while describing specific details for carrying out the invention below.

1 is a block diagram of an air quality measurement system having an automatic calibration algorithm built-in according to an embodiment of the present invention;
Figure 2 is a block diagram showing the control configuration of the sensor device of Figure 1;
3 is a view showing an example of the light scattering measurement unit shown in FIG. 2;
4 is a view showing an example of the alpha-ray and beta-ray ionometry unit shown in FIG. 2;
5 is a control block diagram showing a control configuration of the air quality measuring device of FIG. 2;
6 shows a geometric transformation for generating a discriminant;
7 is a flowchart of an error removing method of the error removing unit of FIG. 5;
8 is a control block diagram showing the control configuration of the calibration unit of FIG. 2;
9 is a flowchart of a control method of the calibration unit.

Terms such as first, second, etc. may be used to describe various elements, but the elements should not be limited by the terms. The above terms are used only for the purpose of distinguishing one component from another. For example, without departing from the scope of the present invention, a first component may be referred to as a second component, and similarly, a second component may also be referred to as a first component. and/or includes a combination of a plurality of related listed items or any of a plurality of related listed items.

When a component is referred to as being “connected” or “connected” to another component, it may be directly connected or connected to the other component, but it is understood that other components may exist in between. it should be On the other hand, when it is said that a certain element is "directly connected" or "directly connected" to another element, it should be understood that the other element does not exist in the middle. Also, when the first and second components on the network are connected or connected, it means that data can be exchanged between the first and second components by wire or wirelessly.

In addition, the suffixes "module" and "part" for the components used in the following description are given simply in consideration of the ease of writing the present specification, and do not impart a particularly important meaning or role by themselves. Accordingly, the terms “module” and “unit” may be used interchangeably.

When these components are implemented in actual applications, two or more components may be combined into one component, or one component may be subdivided into two or more components as needed. The same reference numerals are given to the same or similar components throughout the drawings, and detailed descriptions of the components having the same reference numerals may be omitted instead of being replaced with the descriptions of the above-described components.

Furthermore, the invention encompasses all possible combinations of the embodiments indicated herein. Various embodiments of the present invention are different, but not mutually exclusive. One embodiment of a particular shape, structure, function, and characteristic described herein may be embodied in another embodiment. For example, components mentioned in the first and second embodiments may perform all functions of the first and second embodiments.

Hereinafter, an air quality measuring apparatus according to an embodiment of the present invention will be described in detail with reference to the drawings.

1 is a block diagram of an air quality measurement system having an automatic calibration algorithm built-in according to an embodiment of the present invention, FIG. 2 is a block diagram showing the control configuration of the sensor device of FIG. 1, FIG. 3 is FIG. 2 is a diagram showing an example of the light scattering measurement unit, FIG. 4 is a diagram showing an example of the alpha-ray and beta-ray ionization measurement unit shown in FIG. 2, FIG. 5 is a control showing the control configuration of the air quality measurement device of FIG. 6 is a control block diagram showing a graphic transformation for generating a discriminant, FIG. 7 is a flowchart of an error removing method of the error removing unit of FIG. 5, and FIG. 8 is a control block diagram showing the control configuration of the correcting unit of FIG. , and FIG. 9 is a flowchart of a control method of the calibration unit.

Referring to FIG. 1 , an air quality measurement system having an automatic calibration algorithm embedded therein may include a plurality of sensor devices 20 ( 20-1 to N) and an air quality measurement device.

The plurality of sensor devices 20 may include first to N-th sensor units of the same type for measuring the same object.

2 to 4 , the sensor device 20 may include a light scattering measurement unit 210 , an alpha-ray ionization measurement unit 220 , and a beta-ray ionization measurement unit 230 .

The light scattering measuring unit 210 measures the first current (IL) based on the scattered light that reaches the light receiving unit by scattering the photon beam (hereinafter referred to as 'light') generated from the light source by air, fine dust, moisture, etc. can

3 shows the degree of scattered light such as air, fine dust, and moisture in the light scattering measurement unit 210 . The magnitude of the first current IL may be proportional to the concentration of the suspended material.

The alpha ionization measurement unit 220 may use an alpha ray source emitting alpha rays. In addition, the beta-ray ionization measurement unit 230 may use a beta-ray source emitting beta-rays.

The average distance that ionizing radiation alpha and beta rays travel in air or material is called a range, and the range is the size of the ionizing radiation depending on the energy of the ionizing radiation and the properties of the transmitted material (atomic weight, molecular weight, density, temperature, etc.) may be different.

Ionizing radiation has ionization that ionizes matter, and alpha and beta rays have different degrees of ionization depending on their energy. Also, the degree of ionization may vary depending on the type of ionizing material.

4 is a diagram showing the degree of scattered light such as air, fine dust, and moisture in the alpha-ray ionization measurement unit 220 and the beta-ray ionization measurement unit 230. can be proportional to the concentration of

Referring to FIG. 5 , the air quality measuring device 10 includes a control unit 110 , an error removing unit 120 , a correction unit 130 , a message generating unit 181 , a display unit 183 , a communication unit 185 , and a storage unit. It may include a part 190 .

The error removing unit 120 may remove an erroneous measurement value from among the measurement values of the first to Nth sensor units 20 - 1 to N of the plurality of sensor devices 20 .

The measured values of the first to Nth sensor units 20 - 1 to N may be values calculated by the sensor device 20 or calculated by the air quality measuring device 10 .

For convenience of description, a case in which there are three sensor units of the sensor device 20 will be described with reference to FIGS. 6 and 7 .

6 and 7 , the error removing unit 120 sets the first to third measured values V1, V2, and V3 of the first to third sensor units 20-1 to 3 in the polar coordinate system. (P1, P2, P3) may be converted (S310). The first to third positions P1 to 3 have the first to third measured values V1 to 3 as distances, and angles between adjacent positions are the same. In this embodiment, the angle between them is 120 degrees. In another embodiment, when there are N sensor units, the angle between them may be 360/N angles, and the generated figure may be an N polygon (a polygon having N angles) instead of a triangle.

The error removing unit 120 may convert the first to third positions P1 to 3 into a figure having vertices, that is, a triangle. The error removing unit 120 may obtain first to third lengths d1 to 3 of the side of the generated triangle. If you know the length of both sides and the included angle in a triangle, you can use the formula to find the length of the other side.

The first to third lengths d1 to 3 are affected by the first to third measured values V1 to 3 . Accordingly, the error discriminant expression D may be defined based on the first to third lengths d1 to 3 . When the sensor of the sensor unit is abnormal or dust is attached to the sensor for a long time, at least one of the first to third measured values V1 to 3 indicates a different value. An example of the discriminant (D) for determining this is defined as follows. It will be described that there are N sensor units.

The discriminant (D) is the sum of the differences of the i to jth sides of the N-gon, i is a natural number that is one of 1 to N, and j is a natural number that is one of i to N. can

The error removal unit 120 may calculate an error discrimination value using the above discriminant (D) (S320).

The error removing unit 120 may determine that all of the plurality of sensor units are normal when the error determination value falls within the preset range ( S330 ). If the preset range is found to be within an appropriate value by empirically inferring the discriminant value of the discriminant (D), it can be known whether all data (measured values) are complete or whether there is an error in one or more of the data (measured values).

When the sensor unit is 4 or more, the above discriminant can be applied to all measured values. More preferably, the above discriminant may be applied to some of the plurality of measured values. This is because, when the discriminant value of the discriminant D including the measured value of the specific sensor unit is abnormal and the discriminant value of the discriminant not included is normal, it may be determined that the specific sensor unit causes an error.

The storage unit 190 may store a program for processing and control of the control unit 110 and may perform a function for temporarily storing input or output data.

The storage unit 190 includes a first current IL measured by the light scattering measurement unit 210 , a second current Iα measured by the alpha-ray ionization measurement unit 220 , and a third current measured by the beta-ray measurement unit 230 . Current Iβ can be stored. The storage unit 190 may store the first to third proportional constant set values Q1 to Q3 set by calculating the concentrations of air, fine dust, and moisture in the reference chamber by the concentration determining unit 143 .

The calibration unit 130 may determine a measurement value of one sensor unit 20 .

Referring to FIG. 8 , the calibration unit 130 may include a concentration determining unit 143 and a concentration measuring unit 145 .

The concentration determining unit 143 may determine the first and second concentrations of the first to third currents IL, Iα, and Iβ using the following multiple linear regression equation. The concentration determining unit 143 may receive the first to third currents IL, Iα, and Iβ from the first sensor unit 20 - 1 or the storage unit 190 .

Here, the first concentration may be a concentration of fine dust, and the second concentration may be a concentration of moisture. In addition, it is possible to determine the concentration of air.

Here, IL is the first current, Iα is the second current, Iβ is the third current, A is the concentration value of air, P is the first concentration of fine dust, W is the second concentration of moisture, Ka1, Ka2, and Ka3, respectively is the first proportional constant set value (Q1) indicating the magnitude of the current set in proportion to the air concentration when light, alpha rays, and beta rays are emitted, Kp1, Kp2, and Kp3 are The second set values of proportionality constants (Q2), Kw1, kw2, and Kw3 indicating the magnitudes are the third set values of proportional constants (Q3) indicating the magnitude of the current set in proportion to the moisture concentration when light, alpha rays, and beta rays are emitted.

The average value and measurement error of the first to third proportional constant set values (Q1 to Q3) according to each concentration in a standard chamber that can control the concentrations of air, fine dust, and moisture, respectively, are measured by the concentration measuring unit ( 145), it can be calculated through deep learning operations.

The concentration measuring unit 145 may include a CNN learning processing unit 152 , an LSTM learning processing unit 154 , and an error correcting unit 156 .

The CNN learning processing unit 152 pre-processes the previously input first previous concentrations of fine dust and the second previous concentrations of moisture into a form usable in a prediction model used in a convolutional neural network (CNN), After replacing the missing value generated in the input data with a random value (0 in the experimental environment) and dividing it into a matrix so that it can be processed in the CNN model, a cluster can be assigned according to the similarity of each matrix.

The CNN learning processing unit 152 may learn the CNN neural network using matrix data and clustering results provided in the preprocessing process. Thereafter, the CNN learning processing unit 152 may receive the test data and calculate a CNN classification value.

As a result, the CNN learning processing unit 152 may output the CNN classification values for the first predicted concentration of the fine dust and the second predicted concentration of the moisture.

The LSTM learning processing unit 154 applies the first previous time concentrations of fine dust input before the current time point and the second previous time point concentrations of moisture to the set tanh function, and the third predicted concentration of the fine dust at the current time point and A fourth predicted concentration of the moisture may be output.

The LSTM learning processing unit 154 receives the concentrations of the first previous time points and the second previous time points of water input before the current time into the LSTM neural network, performs prediction, and compares the weights ( weight), it is possible to calculate the LSTM predicted value, that is, the third predicted concentration of the fine dust and the fourth predicted concentration of the moisture.

The error correcting unit 156 calculates the measurement error by comparing the first to fourth predicted concentrations and the first and second concentrations, and corrects the second concentration by the measurement error to obtain the final concentration of the first concentration. It is possible to determine the final concentration (mk) of the fine dust corrected by .

The error correcting unit 156 may calculate the multiple linear regression equation of [Equation 1] described above by performing regression analysis using the first to fourth predicted concentrations and the first and second concentrations.

The error correcting unit 156 may calculate a measurement error by comparing the first to fourth predicted concentrations and the first and second concentrations, respectively, in order to compare the performance of the prediction model.

The error correcting unit 156 may correct the second concentration by the measurement error to determine the final concentration (mk) of the fine dust for which the first concentration is finally corrected.

Also, the error correction unit 156 may correct the multiple linear regression equation shown in Equation 1 by applying the measurement error.

The message generating unit 181 may generate a notification message when the final concentration (mk) of the fine dust measured by the concentration measuring unit 145 is higher than a set reference concentration.

The message generator 181 may generate a warning message corresponding to a warning when the final concentration mk is higher than the reference concentration, and output the final concentration when the final concentration mk is lower than the reference concentration.

The display unit 183 may display at least one of the notification message output from the message generating unit 181 and the final concentration.

The controller 110 may execute an overall control function of the air quality measuring device. The control unit 110 may execute an overall control function of the air quality measuring device by using the program and data stored in the storage unit 190 . The controller 110 may include RAM, ROM, CPU, GPU, and a bus, and the RAM, ROM, CPU, GPU, etc. may be connected to each other through a bus. The control unit 110 may perform any one function of the error removing unit 120 , the correcting unit 130 , and the message generating unit 181 , or may include a block thereof.

The communication unit 185 may transmit at least one of the notification message and the final concentration (mk) to a set mobile terminal, other air quality measurement devices, and data servers. The communication unit 185 may receive the first and second previous concentrations and the first and second previous concentrations from other adjacent air quality measurement devices and data servers.

9 is a flowchart illustrating an operation method of the air quality measurement system according to the present invention.

Referring to FIG. 9 , the light scattering measurement unit 210 scatters a photon beam (hereinafter referred to as 'light') generated from a light source by air, fine dust, moisture, etc. (IL) may be measured (S410).

The alpha ionization measurement unit 220 may measure the second current Iα with respect to the degree of scattered light, such as air, fine dust, and moisture, by using an alpha ray source emitting alpha rays ( S415 ).

The beta-ray ionization measurement unit 230 may measure the third current Iβ with respect to the degree of scattered light, such as air, fine dust, and moisture, using a beta-ray source emitting beta-rays (S420).

The calibration unit 130 includes the first current IL measured by the light scattering measurement unit 210 , the second current Iα measured by the alpha-ray ionization measurement unit 220 , and the third current measured by the beta-ray measurement unit 230 . The current Iβ may be stored (S425).

The calibration unit 130 may determine the first and second concentrations using the first to third currents IL, Iα, and Iβ and the first to third set values of the first to third proportional constants Q1 to Q3 ( S430).

Here, the first concentration may be a concentration of fine dust, and the second concentration may be a concentration of moisture.

The correction unit 130 pre-processes the previously input first previous concentrations of fine dust and second previous concentrations of moisture into a form usable in a prediction model used in a convolutional neural network (CNN), and then CNN classification values for the first predicted concentration of fine dust and the second predicted concentration of moisture may be output (S435).

The correction unit 130 applies the first previous time concentrations of fine dust input before the current time point and the second previous time point concentrations of moisture to the set tanh function to obtain the third predicted concentration of the fine dust at the current time point and the A fourth predicted concentration of moisture may be output (S440).

The calibration unit 130 calculates the measurement error by comparing the first to fourth predicted concentrations and the first and second concentrations, and corrects the second concentration by the measurement error so that the first concentration is finally obtained The corrected final concentration (mk) of the fine dust may be determined (S445).

The calibration unit 130 determines whether the final concentration (mk) of the fine dust measured by the concentration measurement unit 145 is higher than the set reference concentration (S450), and when the final concentration (mk) is higher than the set reference concentration, a notification message ( mm) can be generated (S4550).

In addition, the calibration unit 130 may output the final concentration (mm) when the final concentration (mk) is lower than the reference concentration (S460).

The calibration unit 130 may display at least one of a notification message (mm) and a final concentration (mk).

The calibration unit 130 may transmit at least one of the notification message (mm) and the final concentration (mk) to a set mobile terminal, other air quality measuring devices, and data servers (S465).

The present invention may be implemented in hardware or software. Implementation The present invention can also be implemented as computer-readable codes on a computer-readable recording medium. That is, it may be implemented in the form of a recording medium including instructions executable by a computer. The computer readable medium includes any type of medium in which data that can be read by a computer system is stored. Computer-readable media may include computer storage media and communication storage media. Computer storage media includes all storable media implemented as any method or technology for information storage, such as computer-readable instructions, data structures, program modules, and other data, and includes volatile/nonvolatile/hybrid memory. Whether or not, it is not limited to whether the detachable / non-separable type. Communication storage media includes a modulated data signal or transmission mechanism, such as a carrier wave, any information delivery media, and the like. And functional programs, codes, and code segments for implementing the present invention can be easily inferred by programmers in the technical field to which the present invention pertains.

In addition, although preferred embodiments of the present invention have been illustrated and described above, the present invention is not limited to the specific embodiments described above, and the technical field to which the present invention belongs without departing from the gist of the present invention as claimed in the claims In addition, various modifications are possible by those of ordinary skill in the art, and these modifications should not be individually understood from the technical spirit or prospect of the present invention.

10: air quality measuring device 110: control unit
120: error removal unit 130: correction unit
143: concentration determining unit 145: concentration measuring unit
181: message generating unit 183: display unit
185: communication unit 190: storage unit
20: sensor device 210: light scattering measurement unit
220: alpha ion measurement unit 230: beta ionization measurement unit

Claims (6)

a sensor device including first to N-th sensor units of the same type for measuring the same object; and
The first to N-th sensor unit converts the first to N-th measured values into first to N-th positions having the first to N-th distances in the polar coordinate system, and the first to N-th positions are the vertices. An error removing unit for calculating a discriminant value of a noise discriminant based on the length of each side of , and determining whether all of the first to Nth sensor units are normal based on whether the discriminant value falls within a preset range including an air quality measuring device that
The azimuth angle of the first to Nth positions is an integer multiple of a specific angle, an air quality measurement system with a built-in automatic calibration algorithm. The method of claim 1,
The discriminant is the sum of the differences in the lengths of the i to jth sides of the N-gon, wherein i is any one of 1 to N natural number, and j is any one of i to N natural number,
The discriminant is based on the first to third measurement values of the first to third sensor units, the automatic calibration algorithm is built-in air quality measurement system. The method of claim 1,
The first sensor unit
a light scattering measuring unit measuring a first current based on the scattered light scattered by emitting light to the air in which the suspended particles are filtered in the measuring area;
an alpha ion measurement unit measuring a second current ionized by emitting alpha rays to the air; and
A beta-ray ionization measurement unit for measuring the ionized third current by emitting beta-rays to the air; and
The air quality measuring device calculates the concentrations of air, fine dust and moisture in the reference chamber and based on first to third set values of proportional constants, the first concentration of the fine dust with respect to the first to third currents and A second concentration of moisture is determined, and a deep learning operation is performed on the errors of the first and second concentrations to determine a measurement error for the concentration of fine dust, and the first and second specific concentration data among the concentration data extracted, and the first and second concentration difference values between the first and second specific concentrations of the fine dust included in each of the first and second specific concentration data and the concentration of the fine dust are set in the reference concentration value range If not, the air quality measurement system further comprising a calibration unit for determining the final concentration of the fine dust based on the measurement error and the difference value between the first and second specific concentrations. 4. The method of claim 3,
Further comprising a storage unit for storing the first to third current and the first to third proportional constant set values,
The correction unit,
a concentration determiner configured to determine the first and second concentrations for the first to third currents based on the first to third currents and the first to third set values of proportional constants; and
A concentration measuring unit for measuring the final concentration of the fine dust by deep learning calculation of the measurement error of the first and second concentrations,
The concentration determining unit, the first to third currents are analyzed using the following multiple linear regression equation to calculate the first and second concentrations, the air quality measurement system.

(IL is the first current, Iα is the second current, Iβ is the third current, A is the concentration value of air, P is the first concentration of fine dust, W is the second concentration of moisture, and each of Ka1, Ka2, and Ka3 is The first set of proportional constant values, Kp1, Kp2, Kp3, indicating the magnitude of the current set in proportion to the air concentration when light, alpha ray, and beta ray are emitted 2 Proportional constant set values, Kw1, kw2, and Kw3 are the third set values of proportional constants indicating the magnitude of the current set in proportion to the moisture concentration when light, alpha, and beta rays are emitted) 5. The method of claim 4,
The concentration measuring unit,
Based on the previously input first previous concentrations of fine dust and second previous concentrations of moisture, the first predicted concentration of the fine dust and the second predicted concentration of the moisture with respect to a convolutional neural network (CNN) CNN learning processing unit to output;
The third predicted concentration of the fine dust and the fourth predicted concentration of the moisture at the current time by applying the first previous time concentrations of fine dust and the second previous time concentrations of moisture input before the current time point to the set tanh function LSTM learning processing unit to output; and
The first to fourth predicted concentrations are compared with the first and second concentrations to calculate the measurement error, and by correcting the second concentration by the measurement error, the first concentration of the fine dust is finally corrected. An air quality measurement system comprising an error correction unit for determining the final concentration. The first to N-th measurement values of the first to N-th sensor units of the sensor device having the same first to N-th sensor units for measuring the same object are first to N-th distances in a polar coordinate system. converting to N positions, calculating a discriminant value of a noise discriminant based on the length of each side of an N-gon having the first to Nth positions as a vertex, and determining whether the discriminant value falls within a preset range Including an error removing unit for determining whether all of the first to Nth sensor unit is normal,
The azimuth angle of the first to the Nth position is an integer multiple of a specific angle, an air quality measuring device having.

Images