Sensor correction device, air quality measuring device and correction method thereof

Abstract

The present invention relates to a sensor correction device having a built-in automatic correction algorithm, a sensor correction and air quality measurement device, and a correction method therefor, and more particularly, to a sensor correction device and a correction method therefor, which automatically correct the responsivity of a sensor to improve accuracy. The invention also provides a sensor correction and air quality measurement device and a measurement method thereof, which can accurately measure the fine dust content in the air.

Description

Technical Field

[0001] The present invention relates to a sensor calibration device with a built-in automatic calibration algorithm, a sensor calibration and air quality measuring device and a calibration method thereof, and more specifically, to a sensor calibration device and a calibration method thereof that automatically calibrates sensor responsiveness to improve accuracy, and a sensor calibration and air quality measuring device and a measuring method thereof that accurately measures the amount of fine dust contained in the air. Background Art

[0002] The technology used to remove sensor noise can only be applied when there is reference data that can determine whether the sensor value is correct. In addition, sensor noise reduction technology is often used to improve sensor accuracy rather than real-time noise reduction. Furthermore, existing noise reduction technology cannot be applied in real time, but needs to be implemented in advance during the sensor testing stage (or the testing stage before the product using the sensor is shipped).

[0003] The existing noise removal technology can derive more accurate results as more data is accumulated. However, it is not enough to just have more data. Only when there is data under the maximum number of conditions and environments can noise removal and compensation data be efficiently performed.

[0004] However, there is not always baseline data for all the environments in which sensors are tested and used, and it is difficult to predict all the variables that may occur in the real environment. In addition, it is difficult to get good results in noise removal because it can only cope with the variables that occur during testing, not in real time.

[0005] Due to the above reasons, existing noise reduction technology does not focus on noise reduction, but focuses on sensor calibration, and is therefore difficult to use in actual noise reduction.

[0006] In addition, particulate matter (PM) in the atmosphere is microscopic particles with various chemical properties. In particular, among the fine dust that is directly or indirectly harmful to the human body, particles generated by human industrial activities account for the majority.

[0007] Fine dust is dust particles that are too fine to be seen by the naked eye. Its diameter is less than 10μm. It is classified into PM10 fine dust, PM2.5 ultrafine dust, and PM1.0 ultrafine dust according to its size. PM (Particulate Matter) is "particulate matter (fine particles in solid or liquid state floating in the atmosphere)". PM10 is dust with a particle diameter of less than 10μm, PM2.5 is dust with a diameter of less than 2.5μm, and PM1.0 is dust with a diameter of less than 1.0μm. Ultrafine dust can penetrate into the alveoli and blood vessels, and has a greater impact on health, so it has attracted attention.

[0008] According to the atmospheric pollution engineering test standards, the standard methods for measuring fine dust in the atmospheric environment include the beta-ray absorption method and the weight concentration method.

[0009] The β-ray absorption method is a method for determining the mass concentration of fine dust by measuring the relative intensity of the β-rays absorbed when the β-rays emitted from the β-ray source pass through the dust collected on the air filter. Due to the limitations of the indirect method, there is a disadvantage that the gravimetric concentration method and comparative verification steps are required to derive the accurate mass value.

[0010] The gravimetric concentration method is a method used to measure the mass concentration of PM10 or PM2.5 in the atmosphere. The sample collector is used to collect the dust sample in the atmosphere, and then the weight difference of the filter before and after the collection is calculated as the mass concentration. The gravimetric concentration method has the inconvenience of requiring a long time to collect the sample and it is difficult to grasp the concentration changes during the collection.

[0011] Although it is not the standard method mentioned above, the light scattering method is generally used more frequently. The light scattering method can directly read the results without collecting samples. The light scattering method is a technology that uses the following principle to measure the mass of particulate matter: that is, when light is irradiated to particulate matter (PM) floating in the atmosphere, the light is scattered by the particles. At this time, if the light is irradiated to particulate matter with the same physical properties, the amount of scattered light is proportional to the mass concentration. Among the methods using light scattering, there are turbidity methods that use a light source with three wavelengths to measure the amount of scattered particles, and spectrometer methods that use a single wavelength such as laser or a light source close to a single wavelength generated by an LED to measure the amount of scattered particles.

[0012] The light scattering method uses a weight value (which is used to convert the degree of light scattering caused by dust particles and calculate the number of particles), but there is a problem that its accuracy decreases due to changes in the external environment. In particular, moisture content has the greatest impact on reducing the accuracy of the external environment.

[0013] Recently, research is being conducted to improve the accuracy of moisture content in the air and to measure the exact dust concentration.

[0014] In the field of environmental measurement such as air quality or water quality, many measurement equipment using low-cost and popular sensors have been developed. Recently, with the emergence of topics related to air pollution, the demand for air quality measurement has increased, and the popularity and use of measuring instruments equipped with low-cost sensors (hereinafter referred to as simple measuring instruments) have expanded. Compared with high-priced measurement equipment, the advantages of simple measuring instruments are low price, easy configuration and use, and the disadvantages are relatively low accuracy and short equipment life.

[0015] All expensive measuring equipment used for research and expert use needs to be calibrated according to the national standard traceability system. Similarly, simple measuring instruments also need to go through the calibration process in order to be sold. Expensive measuring equipment needs to be calibrated repeatedly every specific period, while simple measuring instruments purchased and used by general users are only calibrated once during the development period. Therefore, if a simple measuring instrument is used, the measurement performance may sometimes decrease over time. There are many reasons for the performance degradation, among which the performance degradation caused by the life of components that play an important role in sensor measurement, such as heat sinks, laser diodes, photodiodes, and electrochemical cells, is a representative reason.

[0016] <Prior Art Literature>

[0017] (Patent Document 1) Korean Invention Patent Publication No. 10-2018-0041828 Summary of the invention Technical issues

[0018] The object of the present invention is to provide a sensor calibration device and a calibration method thereof with a built-in algorithm for automatically correcting responsiveness, as well as a sensor calibration and air quality measurement device and a measurement method thereof for accurately measuring the amount of fine dust contained in the air.

[0019] Another object of the present invention is to provide an apparatus and method for screening out sensors having errors such as noise in their measurement values ​​using a plurality of sensors.

[0020] In addition, an object of the present invention is to provide an apparatus and method for measuring the amount of moisture and fine dust contained in the air using a light scattering method and a radiation method, and compensating for errors using a deep learning method to easily measure the accurate amount of fine dust. Problem Solution

[0021] In this embodiment, the plurality of sensor units may be sensors of the same type having substantially the same performance or measuring the same type of objects.

[0022] According to an embodiment of the present invention, the sensor calibration method may include: a step of setting a part of the sensor units to a partial activation state, that is, setting a standby group including a first sensor unit among a plurality of sensor units to a standby state, and setting an activation group including a second and third sensor units among the plurality of sensor units to an operating state; a step of maintaining the part of the sensor units in a preset operating period; a step of setting the part of the sensor units to a full activation state after the operating period, that is, setting the standby group and the activation group to an operating state; a step of setting the average value of a correction value obtained by applying a preset first sensor unit correction constant to the actual value of the first sensor unit and the average value of each correction value obtained by applying each preset correction constant to the actual value of each sensor unit in the standby group, i.e., the average value of the standby group, as a comparison reference value after a preset data collection time has passed in the full activation state; a step of comparing the comparison value with a preset correction reference value, scaling the average value of the actual value of the second sensor unit with the comparison reference value as a reference to obtain the comparison value; and a step of setting a value calculated based on the comparison value as a second sensor unit correction constant for the second sensor unit when the comparison value is less than the correction reference value.

[0023] In addition, the sensor calibration method may further include: after the above-mentioned data collection time has passed, the step of setting the above-mentioned first sensor unit to the above-mentioned activation group and the above-mentioned second sensor unit to the above-mentioned standby group; and the step of setting the above-mentioned multiple sensor units to the above-mentioned partial activation state.

[0024] In addition, the sensor calibration method may also include: during the operation period of the above-mentioned partial activation state, a step of determining one of the average value of the correction value after applying the preset second sensor part correction constant to the actual measured value of the above-mentioned second sensor part and the average value of each correction value after applying the respective preset correction constant to the actual measured value of each sensor part of the above-mentioned activation group, that is, the activation group average value, as the final value of the measurement.

[0025] Furthermore, the calibration constant of the second sensor unit may be the inverse of the comparison value.

[0026] Furthermore, the correction value of the second sensor unit may be a value obtained by multiplying the actual measurement value of the second sensor unit by the correction constant.

[0027] In addition, the above-mentioned multiple sensor units have N sensor units, and the sensor calibration method may also include: transforming the 1st measurement value to the Nth measurement value of the above-mentioned 1st sensor unit to the Nth sensor unit into the 1st position to the Nth position having the 1st distance to the Nth distance of the polar coordinate system, calculating the discrimination value of the noise discrimination formula based on the length of each side of the N-sided polygon with the above-mentioned 1st position to the Nth position as vertices, and judging whether all the above-mentioned 1st sensor unit to the Nth sensor unit are normal according to whether the above-mentioned discrimination value is within a preset range.

[0028] The azimuth angles of the first to Nth positions are integer multiples of a specific angle, and N may be 4 or more.

[0029] In addition, the above-mentioned normal judgment can be a preset period, a specific situation, and a situation where the above-mentioned regular value is below a preset error judgment reference value.

[0030] Furthermore, the discriminant is the sum of the differences in lengths of the 1st to jth sides of the N-gon, and the j may be a natural number less than the N and greater than 3.

[0031] Furthermore, the discriminant formula may be based on first to third measurement values ​​of the first sensor, second sensor, and third sensor unit.

[0032] A sensor calibration and air quality measurement method according to an embodiment of the present invention includes the sensor calibration method according to the above embodiment, and the above sensor unit can measure air quality.

[0033] According to an embodiment of the present invention, the sensor calibration and air quality measurement device may include: a plurality of sensor units of the same type, measuring a similar object; and a control unit, wherein the control unit is set to a partially activated state, that is, a standby group having a first sensor unit among the plurality of sensor units is set to a standby state, and an activation group having a second and a third sensor unit among the plurality of sensor units is set to an operating state, and the partially activated state is maintained during a preset operating period, and after the operating period, the partially activated state is set to a fully activated state, that is, the standby group and the activation group are all set to an operating state, and after the preset data collection time has passed in the fully activated state, Then, one of the average value of the correction value after applying the preset first sensor part correction constant to the actual measured value of the above-mentioned first sensor part and the average value of each correction value after applying the respective preset correction constants to the actual measured values ​​of each sensor part of the above-mentioned standby group, i.e., the standby group average value, is set as a comparison reference value, the comparison value is compared with the preset correction reference value, and the average value of the actual measured value of the above-mentioned second sensor part is scaled based on the above-mentioned comparison reference value to obtain the comparison value, and when the above-mentioned comparison value is less than the above-mentioned correction reference value, the value calculated based on the above-mentioned comparison value is set as the second sensor part correction constant for the above-mentioned second sensor part.

[0034] Furthermore, the plurality of sensor units may include first to Nth sensor units of the same type that measure objects of the same type.

[0035] In addition, the above-mentioned control unit can transform the 1st measurement value to the Nth measurement value of the above-mentioned 1st sensor unit to the 1st position to the Nth position having the 1st distance to the Nth distance in a polar coordinate system, calculate the discrimination value of the noise discrimination formula based on the length of each side of the N-sided polygon with the above-mentioned 1st position to the Nth position as vertices, and judge whether the above-mentioned 1st sensor unit to the Nth sensor unit are all normal or not according to whether the above-mentioned discrimination value is within a preset range.

[0036] Furthermore, the azimuth angles of the first to Nth positions may be integer multiples of a specific angle.

[0037] In addition, the above discriminant is the sum of the differences in the lengths of the i-th to j-th sides of the above N-gon, the above i is a natural number from 1 to N, and the above j can be a natural number from i to N.

[0038] Furthermore, the discriminant formula may be based on first to third measurement values ​​of the first sensor, second sensor, and third sensor unit.

[0039] A sensor calibration and air quality measurement device according to an embodiment of the present invention includes the sensor calibration device according to the above embodiment, and the sensor unit can measure air quality.

[0040] In addition, the above-mentioned first sensor unit may include: a light scattering measuring unit, which measures the first current based on scattered light emitted to the air that has filtered the floating microparticles in the above-mentioned measurement area; an α-ray ionization measuring unit, which measures the second current ionized by emitting α-rays to the above-mentioned air; and a β-ray ionization measuring unit, which measures the third current ionized by emitting β-rays to the above-mentioned air.

[0041] In addition, the control unit determines the first concentration of the dust and the second concentration of moisture for the first to third currents according to the first to third proportional constant setting values ​​set based on the concentrations of air, dust and moisture in the calculated reference chamber, and performs deep learning calculations on the errors of the first and second concentrations to determine the measurement error of the dust concentration, and extracts the first and second specific concentration data from the concentration data. If the first and second concentration differences between the first and second specific concentrations of the dust contained in the first and second specific concentration data and the concentration of the dust are not within the preset reference concentration value range, the final concentration of the dust can be determined based on the difference between the first and second specific concentrations and the measurement error.

[0042] Furthermore, a storage unit may be included for storing the first to third currents and the first to third proportional constant setting values.

[0043] In addition, the above-mentioned control unit is equipped with: a concentration determination module, which determines the above-mentioned first and second concentrations for the above-mentioned first to third currents according to the above-mentioned first to third currents and the above-mentioned first to third proportional constant setting values; and a concentration calculation module, which performs deep learning calculations on the measurement errors of the above-mentioned first and second concentrations to determine the final concentration of the above-mentioned dust. The above-mentioned concentration determination module can use the following multiple linear regression equation to analyze the above-mentioned first to third currents and calculate the above-mentioned first and second concentrations.

[0044] I L =k a1 A+k p1 P+k w1 W

[0045] I α =k a2 A+k p2 P+k w2 W

[0046] I β =k a3 A+k p3 P+k w3 W

[0047] 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 dust, W is the second concentration of moisture, Ka1, Ka2, Ka3 are respectively the first proportional constant setting values ​​representing the magnitude of the current set proportional to the air concentration when emitting light, α rays, β rays, Kp1, Kp2, Kp3 are respectively the second proportional constant setting values ​​representing the magnitude of the current set proportional to the air concentration when emitting light, α rays, β rays, Kw1, kw2, Kw3 are respectively the third proportional constant setting values ​​representing the magnitude of the current set proportional to the moisture concentration when emitting light, α rays, β rays.

[0048] In addition, the above-mentioned concentration determination module may include: a CNN learning processing unit, which outputs the first predicted concentration of the above-mentioned dust and the second predicted concentration of the above-mentioned moisture to a convolutional neural network (CNN) based on the first previous concentration of the dust and the second previous concentration of the moisture that were previously input; an LSTM learning processing unit, which applies the first previous concentration of the dust and the second previous concentration of the moisture that were input before the current moment to a preset tanh function, and outputs the third predicted concentration of the above-mentioned dust and the fourth predicted concentration of the above-mentioned moisture at the current moment; and an error compensation unit, which compares the above-mentioned first to fourth predicted concentrations with the above-mentioned first and second concentrations to calculate the above-mentioned measurement error, and compensates the above-mentioned second concentration for the size of the above-mentioned measurement error to determine the final concentration of the above-mentioned dust after the above-mentioned first concentration is finally compensated. Effects of the Invention

[0049] The device and method of the present invention automatically correct the sensor responsiveness to improve the accuracy, and can easily determine whether there is an erroneous sensor among multiple sensors.

[0050] This device and system can perform machine learning / deep learning operations on the first to third currents measured by radiating light, alpha rays and beta rays into the air to determine the final concentration of the dust with the measurement error compensated, thereby being able to accurately measure the concentration of the dust. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 The block diagram is a sensor calibration device for automatically calibrating sensor responsiveness according to an embodiment of the present invention.

[0052] Figure 2 Shown from Figure 1 Definition of the values ​​derived from the sensor section.

[0053] Figure 3 and Figure 4 It is a flow chart of the calibration method of the sensor calibration device.

[0054] Figure 5 Shows the graphical transformation used to generate the discriminant.

[0055] Figure 6 is a module block diagram of a sensor calibration and air quality measurement device according to an embodiment of the present invention.

[0056] Figure 7 It is shown Figure 6 A block diagram of the control structure of the sensor unit.

[0057] Figure 8 It is shown Figure 7 A diagram showing an example of a light scattering measurement unit shown.

[0058] Fig. 9 It is shown Figure 7 FIG. 1 is a diagram showing an example of an α-ray and β-ray ionization measuring unit.

[0059] Fig.10 It is shown Figure 6 A block diagram of the internal structure of the control unit.

[0060] Fig.11 is a diagram showing an operating cycle of a sensor. DETAILED DESCRIPTION

[0061] The terms 1, 2, etc. can be used to describe a variety of components, but the above components should not be limited by the above terms. The purpose of using the above terms is only to distinguish one component from other components. For example, without departing from the scope of the present invention, the first component can be named as the second component, and similarly, the second component can also be named as the first component. The term "and / or" includes a combination of multiple related recorded items or a certain item among multiple related recorded items.

[0062] When a component is "connected" or "connected" to another component, it may be directly connected or connected to the other component, but it should be understood that there may be other components in between. Conversely, when a component is "directly connected" or "directly connected" to another component, it should be understood that there are no other components in between. In addition, the connection or connection between the first component and the second component on the network means that data can be sent and received between the first component and the second component by wire or wirelessly.

[0063] In addition, the suffixes "module" and "unit" used in the following description for the constituent elements are used simply for the convenience of writing this specification, and they themselves do not have a particularly important meaning or function. Therefore, the "module" and "unit" may be used interchangeably.

[0064] When such a constituent element is embodied in an actual application, it is possible that two or more constituent elements are combined into one constituent element, or one constituent element is subdivided into two or more constituent elements as required. The same reference numerals are assigned to the same or similar constituent elements throughout the drawings, and the detailed description of the constituent elements with the same reference numerals may be omitted because they can be replaced by the description of the aforementioned constituent elements.

[0065] In addition, the present invention includes all possible combinations of the embodiments shown in this specification. Although the various embodiments of the present invention are different from each other, they are not mutually exclusive. An embodiment of a specific shape, structure, function and characteristic described in this specification can be implemented by other embodiments. For example, the constituent elements mentioned in the first and second embodiments can perform all the functions of the first and second embodiments.

[0066] Hereinafter, an air quality measuring device according to an embodiment of the present invention will be described in detail with reference to the accompanying drawings.

[0067] Figure 1 is a block diagram of a sensor calibration device for automatically calibrating sensor responsiveness according to an embodiment of the present invention. Figure 2 Shown from Figure 1 Definition of the values ​​derived from the sensor section, 3 and Figure 4 is a flow chart of the calibration method of the sensor calibration device, Figure 5 The graphical transformation used to generate the discriminant value is shown.

[0068] Reference Figure 1 A sensor calibration device for automatically calibrating sensor responsiveness according to an embodiment of the present invention may include a control unit 110 and a sensor device 200 .

[0069] The sensor unit described below may be one sensor or one of a sensor module in which a plurality of sensors are combined. When the sensor unit refers to a certain sensor, the first sensor and the second sensor unit may refer to the first sensor and the second sensor of the same type. When the sensor unit refers to a certain sensor module, the first sensor and the second sensor unit may refer to the first sensor and the second sensor module of the same type. For example, when the sensor unit is a light scattering measurement sensor, the first sensor and the second sensor unit may be the first light scattering measurement sensor and the second light scattering measurement sensor. For another example, when the sensor unit is an air quality measurement sensor module having a light scattering measurement sensor, an α-ray ionization measurement sensor, and a β-ray measurement sensor, the first sensor and the second sensor unit may be the first and the second air quality measurement sensor modules.

[0070] Reference Figure 2 The sensor device 200 may be composed of one 1-1 sensor. The sensor device 200 may be composed of 1-1 to 1-3 sensors of the same type. The sensor device 200 may be composed of the first sensor and the second sensor module of the same type.

[0071] When the sensor device 200 is composed of a first sensor and a second sensor module, the first sensor module is composed of 1-1 to 1-3 sensors of the same type, and the second sensor module is composed of 2-1 to 2-3 sensors of the same type. The 1-1 sensor and the 2-1 sensor are preferably of the same type.

[0072] Reference Figure 2 In this specification, "measured value" may be "measured value" which indicates the actual measurement and export of the sensor, "calibrated value" which is modified by applying calibration constants, etc., "final value" which is the actual value displayed to the user, and "average value" within these specific time windows, but may refer to a part of them according to the context. The measured value may also include "calculated value" which is derived by the sensor module based on the measured value (calibrated value) of each sensor measured for the first time.

[0073] For example, when the sensor device 200 has only one light scattering measurement sensor, the current value measured by the light scattering measurement sensor can be the measured value. When the measured value needs to be compensated and modified, the modified value can be the correction value. The measured value or the correction value can be operated as the corresponding sensor to be finally saved or displayed to the user, and become the final value.

[0074] When the sensor device 200 is composed of multiple light scattering measurement sensors, for example, the first to third light scattering measurement sensors, each sensor outputs the first to third measured values, and the control unit 110 corrects these first to third measured values ​​and converts them into the first to third corrected values. The average value of the first to third corrected values ​​can be used as the final value.

[0075] When the sensor device 200 is a module of multiple different sensors, each sensor unit outputs a series of measured value outputs. The control unit 110 can calibrate these series of measured values, calculate the first to second calculated values ​​from these corrected values, and then combine the first to second calculated values ​​to determine the final value.

[0076] The sensor device 200 may include a plurality of sensor units 210 to 290. The sensor device 200 may include more than two sensor units. The plurality of sensor units may be similar sensor units that measure similar objects. The plurality of sensor units may have substantially the same performance. Thereafter, the plurality of sensor units may have performance such as sensitivity or responsiveness reduced as they are used. Therefore, the sensor needs to be calibrated. In addition, the performance reduction rate may be different for each sensor unit, so this needs to be taken into consideration.

[0077] In order to calibrate the sensor units of the sensor device 200, the control unit 110 operates only a part of the sensor units in a specific cycle, and then operates all the sensor units. After all the sensor units are operated, the control unit 110 can modify (set) the calibration constant of the sensor unit that was operated immediately before according to the measurement value of each sensor unit. The initial calibration constant is preferably 1. Whether the calibration constant is modified or not can be determined based on the measurement value of the sensor unit that was operated immediately before and the measurement value of the sensor unit that was on standby immediately before.

[0078] Hereinafter, a preset time during which only a part of the sensors are operated is referred to as an operation period Ta, and in order to obtain all the measured values, a preset time during which all the sensor units are operated is referred to as a data collection period Tb.

[0079] Reference Figure 2 When the sensor device 200 is initially driven, the control unit 110 may operate all the first sensor unit to the Nth sensor unit (210 to 290) for stabilization of the sensor (S610). For initialization and stabilization of the sensor device 200, the control unit 110 may maintain the plurality of sensor units in the operating state 0 within the stabilization time Tm. The stabilization (S610) step may be optional. The stabilization time Tm may be several hours.

[0080] After the stabilization time Tm has passed, the control unit 110 can use the measurement values ​​of each sensor unit to perform error determination (normality determination) A. The details of the error determination A will be described in detail later.

[0081] The control unit 110 can be set to a partial activation state, that is, the standby group including at least the first sensor unit 210 among multiple sensor units is set to the standby state, and the activation group including at least the second and third sensor units (220, 230) among multiple sensor units is set to the operating state (S615).

[0082] When there are a plurality of sensor units (particularly, three or more), operating two or more sensor units can increase the measurement density. By not operating at least one sensor unit and using the sensors alternately, the responsiveness can be corrected.

[0083] The second and third sensor units (220, 230) may be the same sensor unit. In particular, when the sensor device 200 includes two sensor units, the second and third sensor units (220, 230) are the same sensor unit.

[0084] The control unit 110 may maintain a part of the activation state for a preset operation period Ta (S620). The operation period Ta may be several days.

[0085] The control unit 110 may determine the final value during the operation period Ta of a part of the active state ( S620 ).

[0086] The control unit 110 may determine the activation group average value as the final value to be measured, which is either the average value of the correction value obtained by applying the preset second sensor unit correction constant to the actual measured value of the second sensor unit 220 or the average value of each correction value obtained by applying each preset correction constant to the actual measured value of each sensor unit in the activation group. In terms of measurement accuracy, it is preferable to determine the latter as the final value.

[0087] Before determining the final value, the control unit 110 performs error determination A. The error determination A will be described later.

[0088] After the operation period Ta has passed, the control unit 110 may set the entire group to an active state ( S625 ), that is, may set all of the standby group and the active group to an operating state.

[0089] The control unit 110 may maintain all activation states for a preset data collection time Tb, and determine a final value within the data collection time Tb ( S630 ).

[0090] The final value may be determined based on either the measured value of the standby group sensor unit or the measured value of the active group sensor unit. In terms of accuracy, the measured value of the standby group sensor unit is preferred.

[0091] Before determining the final value, the control unit 110 may perform an error determination A. The error determination A will be described later.

[0092] After the data collection time Tb has passed, the control unit 110 sets one of the average values ​​of the correction values ​​after applying the preset first sensor unit correction constant to the measured value of the first sensor unit 210 and the average values ​​of the correction values ​​after applying the respective preset correction constants to the measured values ​​of the sensor units of the standby group, i.e., the average value of the standby group as the comparison reference value, and calculates the comparison value (S635) by scaling the average value of the measured value of the second sensor unit 220 with the comparison reference value as the reference.

[0093] As an example, the comparison value may be a value obtained by dividing the average value of the measured values ​​of the second sensor unit 220 by the average value of the standby group.

[0094] When the standby group includes two or more sensor units, it is preferable to select an average value of correction values ​​obtained by applying preset correction constants to actual measurement values ​​of the sensor units in the standby group as the first sensor group average value.

[0095] The control unit 110 may compare the comparison value with a preset calibration reference value ( S640 ).

[0096] When the comparison value is equal to or less than the calibration reference value, the control unit 110 may set the value calculated based on the comparison value as the second sensor unit calibration constant for the second sensor unit 220 ( S645 ).

[0097] The second sensor unit calibration constant may be the inverse of the comparison value. The calibration value of the second sensor unit 220 may be a value obtained by multiplying the actual measurement value of the second sensor unit 220 by the calibration constant.

[0098] When the specific end condition ( S660 ) ​​is not met, the control unit 110 may reset the group ( S665 ).

[0099] After the data collection time Tb has elapsed, the control unit 110 may set the first sensor unit 210 to the active group and the second sensor unit 220 to the standby group ( S665 ).

[0100] Thereafter, the control unit 110 may set the plurality of sensor units to a partially activated state ( S615 ).

[0101] The plurality of sensor units may include N sensor units (N is a natural number).

[0102] The control unit 110 may determine whether at least one of the plurality of sensor units should be determined to be abnormal, that is, may determine whether a normal determination condition is met ( S710 ).

[0103] As a basic condition among the conditions for the determination (S710), it is preferred that the number of operating sensor units among the plurality of sensor units is 3 or more.

[0104] The normality determination condition (S710) may include a preset period and a preset specific condition.

[0105] When the stabilization step (S610) is finished, the specific condition may be one of the following: in the final value determination step (S620, S630), and in the judgment of the comparison value and the correction reference value (S640), the comparison value is less than the preset error judgment reference value. The error judgment reference value is a value less than the correction reference value, and the comparison value less than the error judgment reference value is likely to indicate that there is an error in the operating sensor. The comparison value judgment situation may be meaningful when the normality judgment condition step (S710) is not performed in the final value determination step (S620, S630).

[0106] When the normal judgment condition is met, the control unit 110 can make a normal judgment on whether there is an error or abnormality in the multiple sensor units (S715). The control unit 110 can remove the measurement values ​​with errors from the measurement values ​​of the first sensor unit to the Nth sensor unit (210-290) of the multiple sensor devices (200).

[0107] For ease of explanation, refer to Figure 5 , describing the case where the sensor device (200) has three sensor units for operation.

[0108] Reference Figure 5 , the control unit 110 can transform the 1st to 3rd measurement values ​​(V1, V2, V3) of the 1st sensor, the 2nd sensor, and the 3rd sensor unit (210~230) into positions (P1, P2, P3) of the polar coordinate system (S720). The 1st to 3rd positions (P1~3) use the 1st to 3rd measurement values ​​(V1~3) as distances, and the angles between adjacent positions are the same. In the present embodiment, the angle between adjacent positions is 120 degrees. As another example, when there are N sensor units in operation, the angle between adjacent positions can be 360 / N, and the generated figure is not a triangle, but an N-gon (a polygon with N corners).

[0109] The control unit 110 can transform the image into a figure with the first to third positions (P1-3) as vertices, that is, a triangle. The error correction unit (120) can calculate the first to third lengths (d1-3) of the sides of the generated triangle. In a triangle, if the lengths and angles of two sides are known, the length of the remaining side can be used.

[0110] The first to third lengths (d1 to 3) are affected by the first to third measured values ​​(V1 to 3). Therefore, an error discrimination formula D can be defined based on the first to third lengths (d1 to 3). When an abnormality occurs in the sensor of the sensor unit or dust adheres to the sensor for a long time, at least one of the first to third measured values ​​(V1 to 3) shows a different value. As an example of the discrimination formula D for discriminating this, it is defined as follows. This is explained by taking the example of N sensor units.

[0111] The discriminant D is the sum of the differences between the i-th to j-th sides of the N-gon, i is a natural number from 1 to N, j is a natural number from i to N, and can be defined as the following mathematical formula 1.

[0112]

Mathematical formula 1

[0113] The control unit 110 may calculate the error discrimination value using the discriminant equation D ( S725 ).

[0114] The control unit 110 can determine whether there is an error in the measured values ​​of the multiple sensor units according to the error judgment value (S730). When the error judgment value belongs to the preset range, the control unit 110 can determine that the multiple sensor units are all normal. Regarding the preset range, if the judgment value of the discriminant formula D is found within the appropriate value based on experience, it can be known whether the data (measured value) is complete and whether there is more than one error.

[0115] As a result of the error judgment (S730), if there is an error, the control unit 110 can find and remove the measurement value with the error (S735). In order to find the measurement value with the error, the control unit 110 can apply the above-mentioned discriminant formula to a part of the multiple measurement values. This is because when the discriminant value of the discriminant formula D including the measurement value of the specific sensor unit is abnormal, and the discriminant value of the discriminant formula not including the measurement value of the specific sensor unit is normal, it can be determined that the specific sensor unit has an error.

[0116] For example, when N is 4, if the discriminant value of the discriminant equation based on the first to fourth measured values ​​of the first to fourth sensor units is not within the preset range, the control unit 110 may determine that at least one of them is abnormal. Thereafter, when the first discriminant value based on the first to third measured values ​​of the first sensor, the second sensor, and the third sensor units is within the normal range, the control unit 110 may determine that the fourth sensor unit is abnormal.

[0117] The control unit 110 may determine whether the error is temporary (S740). For example, if an error continues to occur in the measurement value of a specific sensor unit for a predetermined time or longer, it may be determined that the error is not temporary.

[0118] When the error is temporary, the control unit 110 may treat it as noise and stabilize the corresponding sensor unit by temporarily stopping the operation and then restarting the operation ( S750 ).

[0119] When the error is not temporary, the control unit 110 may stop the operation of the corresponding sensor unit (S745). The control unit 110 may notify the manager of the abnormality of the sensor or record it in a log file.

[0120] Figure 6 is a module block diagram of a sensor calibration and air quality measurement device according to an embodiment of the present invention. Figure 7 It is shown Figure 6 The module block diagram of the control structure of the sensor part, Figure 8 It is shown Figure 7 FIG. 1 is a diagram showing an example of a light scattering measurement unit, Fig. 9 It is shown Figure 7 FIG. 1 is a diagram showing an example of an α-ray and β-ray ionization measuring unit, Fig.10 It is shown Figure 6 A block diagram of the internal structure of the control unit, Fig.11 is a diagram showing an operating cycle of a sensor.

[0121] Reference Figure 1 to Figure 5 The description of the components of this embodiment can be replaced by the description of the components with the same names and / or reference numerals as described above, and thus the description may be omitted.

[0122] Reference Figure 6 The sensor calibration and air quality measurement device may include a control unit 110 , a storage unit 190 and a sensor device 200 .

[0123] The sensor device 200 may include first to Nth sensor units 210 to 290 .

[0124] Reference Figure 7 The first sensor unit 210 may include at least one of a light scattering measurement sensor 211 , an α-ray ionization measurement sensor 212 , and a β-ray ionization measurement sensor 213 .

[0125] The light scattering measuring sensor 211 can measure the actual value, ie, the light scattering current IL, based on the scattered light that reaches the light receiving part when the photon beam (hereinafter referred to as "light") generated by the light source is scattered by air, dust, moisture, etc.

[0126] Figure 8 The first current IL shows the degree of light scattered by air, dust, water, etc. in the light scattering measurement sensor 211. The magnitude of the first current IL may be proportional to the concentration of suspended matter.

[0127] When the sensor device 200 includes only one light scattering measurement sensor 211 , generally, the control unit 110 obtains a first correction value by applying a first correction constant to a first measured value, and determines the air quality index as a final value after appropriately manipulating the first correction value.

[0128] The α-ray ionization measurement sensor 212 can use an α-ray source that radiates α-rays Rα. Also, the β-ray ionization measurement sensor 213 can use a β-ray source that radiates β-rays Rβ.

[0129] The average distance that ionizing radiation, i.e., α-rays and β-rays, travel in air or matter is defined as the range. The range may vary depending on the energy of the ionizing radiation and the properties of the material it passes through (atomic weight, molecular weight, density, temperature, etc.).

[0130] Ionizing radiation has an ionizing effect that ionizes matter. The degree of ionization of α-rays and β-rays varies depending on their energy. In addition, the degree of ionization may also vary depending on the type of matter being ionized.

[0131] Fig. 9 The degree of scattered light caused by air, dust, moisture, etc. in the α-ray ionization measuring sensor 212 and the β-ray ionization measuring sensor 213 is shown. The magnitude of each first measured value, namely the α / β ionization current (Iα, Iβ), is proportional to the concentration of suspended matter.

[0132] The storage unit 190 may store a program for processing and control of the control unit 110 and may also perform a function of temporarily storing input or output data.

[0133] The storage unit 190 can store the first current IL measured by the light scattering measurement unit 210, the second current Iα measured by the α-ray ionization measurement unit 220, and the third current Iβ measured by the β-ray measurement unit 230. The storage unit 190 can store the first to third proportional constant setting values ​​Q1 to Q3 set by the concentration determination unit 143 calculating the concentrations of air, dust, and water in the reference chamber.

[0134] The control unit 110 can perform the overall control function of the air quality measuring device. The control unit 110 can perform the overall control function of the air quality measuring device using the program and data stored in the storage unit 190. The control unit 110 may include RAM, ROM, CPU, GPU, and a hub, and the RAM, ROM, CPU, GPU, etc. may be connected to each other through a hub.

[0135] Next, a method for deriving a final value representing air quality is described when the sensor device 200 includes a first sensor unit 210, and the first sensor unit 210 includes a first sensor, a second sensor, and a third sensor (a light scattering measurement sensor 211, an α-ray ionization measurement sensor 212, and a β-ray ionization measurement sensor (213)).

[0136] Reference Fig.10 , the control unit 110 may include a concentration determination module 140 and a concentration calculation module 150 .

[0137] The concentration determination module 140 may determine the first and second concentrations by applying the following multiple linear regression equation to the first to third currents (IL, Iα, Iβ). The concentration determination module 140 may receive the first to third currents (IL, Iα, Iβ) from the first sensor unit 210 or the storage unit 190 .

[0138] Here, the first concentration is the concentration of fine dust, and the second concentration may be the concentration of water. Alternatively, the concentration of air may be determined.

[0139]

Mathematical formula 2

[0140] I L =k a1 A+k p1 P+k w1 W

[0141] I α =k a2 A+k p2 P+k w2 W

[0142] Iβ =k a3 A+k p3 P+k w3 W

[0143] Here, IL is the first current, Iα is the second current, Iβ is the third current, A is the air concentration value, P is the first concentration of dust, W is the second concentration of moisture, Ka1, Ka2, Ka3 are respectively the first proportional constant setting value Q1 representing the magnitude of the current set in proportion to the air concentration when emitting light, α rays, β rays, Kp1, Kp2, Kp3 are respectively the second proportional constant setting value Q2 representing the magnitude of the current set in proportion to the air concentration when emitting light, α rays, β rays, Kw1, kw2, Kw3 are respectively the third proportional constant setting value Q3 representing the magnitude of the current set in proportion to the moisture concentration when emitting light, α rays, β rays.

[0144] The first to third proportional constant setting values ​​Q1 to Q3 may be obtained by various methods. For example, the concentration calculation module 150 may calculate the average value and measurement error of the first to third proportional constant setting values ​​(Q1 to Q3) based on the concentrations in the standard chamber that can control the concentrations of air, dust and water, respectively, by deep learning operation.

[0145] The concentration calculation module 150 may include a CNN learning processing unit 152 , an LSTM learning processing unit 154 , and an error compensation unit 156 .

[0146] The CNN learning processing unit 152 can pre-process the previously input first previous concentrations of dust and the second previous concentrations of water into a form that can be used in the prediction model used by the convolutional neural network (CNN), replace the missing values ​​(Missing Value) generated in the input data with an arbitrary value (0 in the experimental environment), and then divide it into a matrix form for processing in the CNN model, and then assign clusters according to the similarity of each matrix.

[0147] The CNN learning processing unit 152 can learn the CNN neural network using the matrix data and clustering results provided in the preprocessing process. Afterwards, the CNN learning processing unit 152 can receive the input of the test data to calculate the CNN classification value.

[0148] As a result, the CNN learning processing unit 152 can output CNN classification values ​​for the first predicted concentration of the fine dust and the second predicted concentration of the water.

[0149] The LSTM learning processing unit 154 may apply the first previous concentrations of dust and the second previous concentrations of water input before the current moment to a preset tanh function to output the third predicted concentration of the dust and the fourth predicted concentration of water at the current moment.

[0150] The LSTM learning processing unit 154 receives the first previous moment concentration of the dust and the second previous moment concentration of the water input into the LSTM neural network before the current moment to perform prediction, and performs learning of adjusting the weights of the neural network by comparing with the actual values, and calculates the LSTM prediction value, i.e., the third predicted concentration of the above-mentioned dust and the fourth predicted concentration of the above-mentioned water.

[0151] The error compensation unit 156 compares the first to fourth predicted concentrations with the first and second concentrations, calculates the measurement error, and compensates the second concentration for the measurement error to determine a final concentration mk of the fine dust after the first concentration is finally compensated.

[0152] The error compensation unit 156 may perform regression analysis using the first to fourth predicted concentrations and the first and second concentrations to calculate the multivariate linear regression equation of the mathematical formula 2.

[0153] In order to compare the performance of the prediction model, the error compensation unit 156 may compare each of the first to fourth predicted concentrations with the first and second concentrations to calculate the measurement error.

[0154] The error compensation unit 156 may compensate the second concentration for the measurement error to obtain a final concentration mk of the dust after the first concentration is finally compensated.

[0155] In addition, the error compensation unit 156 may apply the above-mentioned measurement error to correct the multivariate linear regression equation shown in Mathematical Formula 2.

[0156] The following describes the embodiments by dividing them into different types according to the number of sensor units.

[0157] When the sensor unit is a sensor, the number of sensors is N. Assuming that the number of sensors is N, there are a first sensor, a second sensor, ..., an Nth sensor, and the operation state of each sensor can be defined as a 0th operation state, a 1st operation state, a 2nd operation state, ..., an Nth operation state, etc. The 0th operation state is an operation mode in which all sensors operate, and the 1st operation state indicates a state in which the 1st sensor is not operating (standby mode) and the remaining sensors are operating.

[0158] [Example 1] 2 sensors (simple cross-use sensors)

[0159] The following description will take the case where there are two sensors as an example. The sensor device 200 may include a first sensor unit 210 and a second sensor unit 220 .

[0160] Reference Figure 7 The first sensor unit 210 and the second sensor unit 220 may each include a light scattering measurement sensor. In this embodiment, it is assumed that the light scattering measurement sensor 211 includes a first light scattering measurement sensor and a second light scattering measurement sensor.

[0161] In this embodiment, each operation state is as follows: Hereinafter, the first light scattering measurement sensor and the second light scattering measurement sensor are referred to as the first sensor and the second sensor.

[0162] -0th action state: 1st sensor operation mode & 2nd sensor operation mode

[0163] -1st operating state: 1st sensor standby mode & 2nd sensor operating mode

[0164] - Second operating state: 1st sensor operation mode & 2nd sensor standby mode

[0165] When using only two sensors, the responsiveness can be calibrated by using the sensors alternately. The responsiveness of the sensors can be calibrated by operating in state 0.

[0166] Reference Fig.11 During initial use, the control unit 110 may maintain the 0th action state for the stabilization time Tm (eg, 1 hour) required for sensor calibration and stabilization of the air quality measurement device.

[0167] After stabilization is completed, the control unit 110 can switch to the first operating state, that is, the partially activated state. The initial calibration constant is preferably 1. Therefore, in this interval, the control unit 110 can obtain the final value based on the actual measurement value of the second sensor.

[0168] The control unit 110 may maintain a portion of the activation state for a preset operation period Ta (eg, 7 days).

[0169] After the operation period Ta has elapsed, the control unit 110 may be set to the 0th operation state, that is, to make both the first sensor and the second sensor enter the operation state.

[0170] The control unit 110 maintains all activation states for a preset data collection time Tb (eg, 1 hour), and determines a final value within the data collection time Tb.

[0171] The final value in the zeroth operation state may be determined based on one of the measurement value of the first sensor, the measurement value of the second sensor, and the average value of the measurement values ​​of the first sensor and the second sensor. From the perspective of accuracy, the measurement value of the first sensor is preferably used.

[0172] After the data collection time Tb has passed, the control unit 110 may set the average value bar(S1)Xk1 of the first correction value obtained by applying the preset first sensor correction constant k1 to the first measured value S1 of the first sensor as the comparison reference value. The control unit 110 may calculate a comparison value d=bar(S2) / bar(S1) obtained by scaling the average value bar(S2) of the measured value S2 of the second sensor unit 220 with respect to the comparison reference value bar(S1).

[0173] The control unit 110 may compare the comparison value d with a preset correction reference value (eg, 0.9).

[0174] When the comparison value d is greater than the calibration reference value, the control unit 110 may maintain the previous second sensor calibration constant k2 ′.

[0175] When the comparison value d is less than the calibration reference value, the control unit 110 may set the value calculated based on the comparison value d as the second sensor calibration constant k2 for the second sensor. In this case, the second sensor calibration constant k2 may be the reciprocal of the comparison value. For this, refer to the following mathematical formula 3.

[0176]

Mathematical formula 3

[0177]

[0178] In the next first operating state, the correction value of the second sensor may be a value obtained by multiplying the actual measurement value of the second sensor by the second sensor correction constant k2.

[0179] After the calibration of the second sensor is completed, the control unit 110 switches to the second operating state to operate the air quality measuring device for 7 days, then switches to the 0 operating state again to collect data for 1 hour, and then applies Mathematical Formula 4 to the first sensor. For the meaning of each symbol, refer to Mathematical Formula 3.

[0180]

Mathematical formula 4

[0181]

[0182] The following table 1 shows that the dust concentration is maintained at 100 μg / m 3 An example of using two sensors and applying an algorithm inside a cavity.

[0183]

Table 1

[0184] [Example 2] 3 sensors

[0185] The following description will be given by taking the case where there are three sensors as an example. Please refer to the description when there are two sensors.

[0186] Assume that the sensor device 200 of the sensor calibration and air quality measurement device includes a first sensor unit 210 , a second sensor unit 220 , and a third sensor unit 230 , each of which includes a light scattering measurement sensor 211 , an α-ray ionization measurement sensor 212 , and a β-ray ionization measurement sensor 213 .

[0187] The first sensor unit 210, the second sensor unit 220, and the third sensor unit 230 output the first air concentration measured value, the second air concentration measured value, and the third air concentration measured value respectively through the concentration determination module 140 and the concentration calculation module 150. In the following, the first sensor unit 210, the second sensor unit 220, and the third sensor unit 230 are represented as the first sensor, the second sensor, and the third sensor, and the first air concentration measured value, the second air concentration measured value, and the third air concentration measured value are represented as the first measured value S1, the second measured value S2, and the third measured value S3, and each correction value is the product of the first sensor k1, the second sensor k2, and the third sensor correction constant k3 and the first measured value S1, the second measured value S2, and the third measured value S3 respectively. Previously, the first sensor correction constant, the second sensor correction constant, and the third sensor correction constant were represented as k1' to k3', respectively.

[0188] In this embodiment, the operation status is as follows.

[0189] -0th action state: 1st sensor is running & 2nd sensor is running & 3rd sensor is running

[0190] -1st operating state: 1st sensor standby & 2nd sensor operating & 3rd sensor operating

[0191] - Second operating state: 1st sensor operating & 2nd sensor standby & 3rd sensor operating

[0192] -3rd action state: 1st sensor operating & 2nd sensor operating & 3rd sensor standby

[0193] The control unit 110 may switch the operation of the first sensor, the second sensor, and the third sensor in the order of the first action state, the 0th action state, the second action state, the 0th action state, the 3rd action state, the 0th action state, the first action state, and the 0th action state. The final value in the order of each action state may be defined as follows.

[0194] The first action state: {k_2S_2+k_3S_3}over 2

[0195] Action state 0: {k_2S_2+k_3S_3}over 2 {k_1S_1+k_2S_2+k_3S_3}over 3

[0196] Second action state: {k_3S_3+k_1S_1}over 2

[0197] The calibration constants for each sensor can be defined as follows.

[0198]

Mathematical formula 5

[0199]

[0200] i: Sensor number to be calibrated

[0201] j: Immediately preceding action state number (standby mode sensor number)

[0202] k i : The calibration constant to be set for the sensor to be calibrated

[0203] k i ”: The calibration constant of the sensor to be calibrated immediately before

[0204] k j : Calibration constant of sensor j

[0205] The average value of the actual measurement value of sensor i in the 0th operation state

[0206] The average value of the actual measurement value of sensor j in the 0th operation state

[0207] For example, when the 1st action state is converted to the 0th action state, i can be 2 or 3, and j can be 1.

[0208] [Example 3]: 4 or more sensors

[0209] The following description will be given by taking the case where there are four or more sensors as an example, and the descriptions for the case where there are two sensors and three sensors will be referred to.

[0210] The control unit 110 can determine whether the measured values ​​or the average value of the measured values ​​of the first sensor unit to the Nth sensor unit (210 to 290) are within the range based on Figure 4 When an error occurs, the control unit 110 eliminates the corresponding error, determines the final value, or performs correction constant compensation.

[0211] Reference Figure 6The sensor calibration and air quality measurement device may further include at least one of a message generating unit 181 , a display unit 183 and a communication unit 185 .

[0212] When the final concentration (final value) of the fine dust measured by the control unit 110 is higher than a preset reference concentration, the message generation unit 181 may generate a notification message.

[0213] When the final concentration is higher than the reference concentration, the message generation unit 181 generates a notification message corresponding to a warning, and when the final concentration is lower than the reference concentration, the final concentration may be output.

[0214] The display unit 183 may display at least one of the notification message output by the message generating unit 181 and the final concentration.

[0215] The communication unit 185 may send at least one of the notification message and the final concentration to a preset mobile terminal, other multiple air quality measuring devices, and multiple data servers. The communication unit 185 may receive the first previous concentration and the second previous concentration and the first previous time concentration and the second previous time concentration from other multiple adjacent air quality measuring devices and multiple data servers.

[0216] In this embodiment, the sensor calibration device or the sensor calibration and air quality measurement device described above calculates the measurement value internally, but the actual measurement value may be transmitted to an external server via the communication unit 185 to calculate the final value on the external server.

[0217] The present invention described above can be implemented in hardware or software. The present invention can be implemented as a computer-readable code on a computer-readable recording medium. That is, it can be implemented in the form of a recording medium containing commands that can be executed by a computer. Computer-readable media include all kinds of media for storing data that can be read by a computer system. Computer-readable media may include computer storage media and communication storage media. Computer storage media include all storable media implemented by any method or technology for storing information such as computer-readable commands, data structures, program modules and other data, not limited to volatile / non-volatile / composite memory, separate / non-separable, etc. Communication storage media include coordinated data signals or transmission mechanisms such as carrier waves, and any information transmission media. In addition, the functional programs, codes, code segments, etc. used to implement the present invention can be easily inferred by programmers in the technical field to which the present invention belongs.

[0218] In addition, the preferred embodiments of the present invention are illustrated and described above, but the present invention is not limited to the above-mentioned specific embodiments. Various modified implementations can be performed by ordinary technicians in the technical field to which the present invention belongs without departing from the main purpose of the present invention as requested by the claims. Such modified implementations should not be understood as being different from the technical ideas or prospects of the present invention.

[0219] <Description of Reference Numerals>

[0220] 110: Control unit 120: Error correction module

[0221] 130: Correction module 140: Concentration determination module

[0222] 150: concentration calculation module 200: sensor device

[0223] 210 to 290: 1st sensor unit to Nth sensor unit 211: Light scattering measurement sensor

[0224] 212: α-ray ionization measurement sensor 213: β-ray ionization measurement sensor

Claims

1. A sensor calibration method, characterized in that: include: a step of setting a part of the sensor units to an activated state, i.e., setting a standby group including a first sensor unit (210) among the plurality of sensor units to a standby state, and setting an active group including a second and a third sensor unit among the plurality of sensor units to an operating state; The step of maintaining the activation state of the above part during a preset operation period; After the operation period, a step of setting all the groups to an active state, that is, setting all the groups to an operating state; After a preset data collection time has elapsed in all the activation states, a step of setting as a comparison reference value an average value of a correction value obtained by applying a preset first sensor unit correction constant to the actual measured value of the first sensor unit and an average value of respective correction values ​​obtained by applying respective preset correction constants to the actual measured values ​​of the sensor units of the standby group, i.e., an average value of the standby group; The step of comparing the comparison value with a preset calibration reference value, scaling the average value of the actual measurement value of the second sensor unit based on the comparison reference value to obtain the comparison value; and A step of setting a value calculated based on the comparison value as a second sensor unit calibration constant for the second sensor unit when the comparison value is equal to or less than the correction reference value.

2. The sensor calibration method according to claim 1, characterized in that: Also includes: After the data collection time has elapsed, the first sensor unit is set as the active group and the second sensor unit is set as the standby group; and The step of setting the plurality of sensor units to the partially activated state.

3. The sensor calibration method according to claim 1, characterized in that: Also includes: During the operation period of the above-mentioned partial activation state, a step is performed in which one of the average value of the correction value after applying the preset second sensor part correction constant to the actual measured value of the above-mentioned second sensor part and the average value of each correction value after applying the respective preset correction constant to the actual measured value of each sensor part of the above-mentioned activation group, that is, the activation group average value, is determined as the final value of the measurement.

4. The sensor calibration method according to claim 1, characterized in that: The plurality of sensor units include N sensor units, The sensor calibration method further includes: converting the first to Nth measured values ​​of the first to Nth sensor units into first to Nth positions having first to Nth distances in a polar coordinate system, calculating a discrimination value of a noise discrimination equation based on the lengths of each side of an N-sided polygon having the first to Nth positions as vertices, and judging whether all of the first to Nth sensor units are normal based on whether the discrimination value is within a preset range; The azimuth angles of the first to Nth positions are integer multiples of a specific angle, and N is 4 or more.

5. A sensor calibration and air quality measurement method, characterized in that: The sensor calibration method according to any one of claims 1 to 4, The sensor unit measures air quality.

6. A sensor calibration device, characterized in that: include: Multiple sensor units of the same type are used to measure objects of the same type; and Control Department, The above control unit, The sensor units are set to a partially activated state, that is, a standby group including a first sensor unit among the plurality of sensor units is set to a standby state, and an active group including a second and a third sensor unit among the plurality of sensor units is set to an operating state. Keep the above part activated during the preset operation period. After the above operation period, the state is set to all active, that is, the above standby group and the above active group are all set to the operation state. After a preset data collection time has elapsed in all the above-mentioned activation states, the average value of the correction value obtained by applying the preset first sensor unit correction constant to the actual measured value of the above-mentioned first sensor unit and the average value of each correction value obtained by applying each preset correction constant to the actual measured value of each sensor unit of the above-mentioned standby group, i.e., the standby group average value, is set as a comparison reference value. The comparison value is compared with a preset calibration reference value, and the average value of the actual measurement value of the second sensor unit is scaled based on the comparison reference value to obtain the comparison value. When the comparison value is equal to or less than the correction reference value, a value calculated based on the comparison value is set as a second sensor unit calibration constant for the second sensor unit.

Images