KR20220036737A - Indoor air quality monitoring system and air quality prediction method using the same - Google Patents

Abstract

According to the present invention, disclosed are an indoor air quality monitoring system and an air quality level prediction method using the same. The indoor air quality monitoring system of the present embodiment is an air quality monitoring system that is installed in a facility of an air pollution vulnerable class and monitors a change in indoor air quality, and comprises: a first information measurement unit having a sensor module for detecting a plurality of harmful substances and measuring the air quality inside the facility in real time; a big data storage unit that cumulatively stores and manages air quality information inside the facility; a second information collection unit for collecting past air quality change information from the big data storage unit; an air quality level prediction unit that predicts a change in air quality measured by the first information measurement unit on the basis of the second information collected by the second information collection unit through machine learning and determines a risk level; an air quality information providing unit that provides information on a change in air quality predicted by the air quality level prediction unit and a dangerous level to a user interface; and a device driving unit that drives a ventilation device or a purification device before air pollution reaches a dangerous level according to a prediction result of the air quality level prediction unit. Air pollution inside a facility is prevented from reaching critical values.

Description

Indoor air quality monitoring system and air quality prediction method using the same

The present invention discloses an indoor air quality monitoring system and a method for predicting an air quality level through the same.

Since most city dwellers spend most of their day in buildings, indoor air quality can have a direct impact on the health of city dwellers. In other words, modern people suffering from various respiratory diseases due to exposure to contaminated air continuously circulated through artificial facilities in a limited space, or living for a long time in a closed space without ventilation facilities, inhaling various pollutants harmful to the human body. This is increasing.

Recently, as indoor environmental problems have emerged as a major issue for public health, the need for real-time management of major pollutants is emerging. In particular, according to the economic level and the development of science, there is an increasing number of cases of complaining of building syndrome symptoms from residents living in indoor spaces, such as various pollutants being emitted from complex and diverse household items and building interior materials.

Indoor air pollution is caused by various causes, and some devices such as air purifiers are used to improve the indoor air quality. In addition, since there is no measurement sensor attached to measure the numerical value, it was difficult for occupants to check the amount of pollutants numerically, making it impossible to effectively manage air quality improvement.

To this end, recently, various types of air pollution detection systems and devices that automatically measure indoor air quality and inform air pollution status along with numerical values to induce ventilation or automatically drive air purification devices according to the degree of pollution have been developed and disseminated. is becoming As an example, Korean Patent No. 10-2008855 introduces 'indoor/outdoor air quality measurement and ventilation control sharing platform service providing system', and Korean Patent No. 10-1591735 introduces 'a method of providing pollution information through prediction of indoor air quality' are doing

The air purification system of the preceding documents provides air pollution information to the user by measuring or predicting indoor or outdoor air pollution in real time, so that the user can take action in advance. However, the air purification system of the prior literature predicts air pollution based on air quality measured in real time, and the data base for prediction is weak, so the accuracy of the prediction information is poor. Therefore, the progress of air pollution can be variously progressed according to region, time, type of facility, etc. It is necessary to improve the accuracy of prediction information by synthesizing these air pollution factors and predicting air pollution.

On the other hand, in the case of an air purification system inside a general facility, adults living indoors instinctively detect air pollution before it reaches a critical level based on information provided by the air purification system or based on experience, and use the purification system in advance. Alternatively, the ventilation system may be activated. However, in a facility where the vulnerable group against air pollution resides, there are cases where the purifier or ventilation system cannot be operated in time when the manager or guardian is absent. In addition, for the safety-vulnerable class, even if ventilation or purification is performed when air pollution reaches the threshold, it can already pose a fatal risk when the air pollution reaches the threshold, so It is necessary to inform the situation of contamination and activate the purification or ventilation system.

The present invention has been proposed in order to solve the above problems, and the problem to be solved by the present invention is to monitor the harmful components of the indoor air in real time, and to inform in advance of the increase in the specific gravity of the harmful components through prediction so that it can be prevented in a timely manner. It is to provide an indoor air quality monitoring system and a method for predicting the air quality level through it.

In addition, the present invention predicts and provides in advance the timing of reaching the threshold by analyzing the change of harmful components in the past time-series, so that the ventilation device or the purification device can be driven before the air pollution state reaches the threshold. An object of the present invention is to provide an indoor air quality monitoring system and an air quality level prediction method through it.

In addition, the present invention provides an indoor air quality monitoring system that automatically purifies the air in advance even in the absence of a manager or guardian in a facility used by the safety-vulnerable class, such as a facility for infants and toddlers, a facility for the disabled, or a nursing facility, and a method for predicting the air quality level through it The task is to provide

In addition, the present invention predicts the time to reach the dangerous level of air quality based on data provided from them in conjunction with an already installed and operating air purification system or ventilation system, and drives the ventilation device or purification device in advance, so that the air quality is reduced to a dangerous level It is an object to provide an indoor air quality monitoring system that does not reach

The indoor air quality monitoring system of this embodiment for solving the above problems is an air quality monitoring system that is installed in a facility of the air pollution vulnerable group to monitor changes in indoor air quality, and includes a sensor module for detecting a plurality of harmful substances A first information measuring unit that measures the air quality inside the facility in real time, a big data storage unit that accumulatively stores and manages air quality information inside the facility, and the big data storage unit to collect past air quality change information A second information collection unit, an air quality level prediction unit that predicts a change in air quality measured by the first information measurement unit based on the second information collected by the second information collection unit through machine learning and determines a risk level; An air quality information providing unit that provides information on a change in air quality and a dangerous level predicted by the air quality level prediction unit to a user interface, and a ventilation device or purification system before air pollution reaches a dangerous level according to the prediction result of the air quality level prediction unit Including a device driver that drives the device, prevents air pollution inside the facility from reaching a threshold.

In addition, in this embodiment, the big data storage unit further stores and manages facility information, the second information collection unit further collects the facility information, and the air quality level prediction unit includes the past air quality change information and the facility information. and may be configured to predict an air quality level based on the information.

In addition, in this embodiment, the facility information may include space size and past ventilation time information of the facility.

In addition, in this embodiment, a weather information collecting unit for collecting meteorological information from an external agency; further comprising, wherein the device driving unit includes the air quality level predicted by the air quality level prediction unit and weather information collected from the weather information By comparison, it may be configured to selectively drive a ventilation device or a purification device.

In addition, the air quality level prediction method of this embodiment for solving the above problems is an air quality level prediction method for predicting changes in air quality and a dangerous level through an air quality monitoring system installed in a facility of the air pollution vulnerable class, ( a) measuring real-time indoor air quality, (b) extracting indoor air quality change information from the big data storage unit, (c) targeting the past air quality change information extracted in step (b) Predicting the change in air quality and the risk level measured in step (a) through machine learning, (d) providing information on the change in air quality and the risk level predicted in step (c) to a user interface, and , (e) in response to the dangerous level of air quality predicted in step (c), before the air quality reaches the dangerous level, operating a ventilation device or a purifier.

In addition, in this embodiment, the step (b) further extracts facility information including facility space size and past ventilation information of the facility from the big data storage unit, and the step (c) is the past air quality change The level of air quality can be predicted by machine learning on the information and the facility information.

In addition, in this embodiment, (f) receiving the air quality information measured in step (a) and the ventilation information of step (e) may further include the step of constructing learning data.

The indoor air quality monitoring system of this embodiment and the air quality level prediction method through it are configured to ventilate or purify the air before reaching a threshold of air pollution measured in real time by using the past air quality change trend and facility information. In particular, it is possible to maintain a high-quality air condition for facilities in which the vulnerable live, such as facilities for infants, children with disabilities, and nursing homes.

In addition, the present invention can operate the ventilation device or the purification device in advance by predicting the time to reach the dangerous level of air quality based on the data provided from the air purification system or the ventilation system that is already installed and operating.

1 is a conceptual diagram showing an implementation example of an indoor air quality monitoring system according to this embodiment;
2 is a block diagram illustrating the main configuration of an indoor air quality monitoring system according to the present embodiment;
3 is a block diagram illustrating a sensor module of a first information storage unit, which is a main part of FIG. 2;
4 is a flowchart showing the air quality level prediction process according to the present embodiment;
5 is a data table illustrating information measured by the first information measuring unit according to the present embodiment;
6 is a diagram illustrating structured data stored in a big data storage unit according to the present embodiment;
7 is a data table exemplifying source data and structured data information of the big data storage unit according to the present embodiment;
8 is a data table illustrating an indoor air quality level prediction result according to the present embodiment.

The present invention and the technical problems achieved by the practice of the present invention will be made clear by the preferred embodiments described below. Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings.

It should be understood that the differences between the embodiments described below are not mutually exclusive. That is, without departing from the spirit and scope of the present invention, the specific shapes, structures and characteristics described may be implemented in other embodiments with respect to one embodiment, and the location of individual components within each disclosed embodiment Or it should be understood that the arrangement may be changed.

1 is a conceptual diagram showing an implementation example of an indoor air quality monitoring system according to this embodiment, FIG. 2 is a block diagram illustrating a main configuration of an indoor air quality monitoring system according to this embodiment, and FIG. 3 is a main part of FIG. It is a block diagram illustrating the sensor module of the first information storage unit.

First, referring to FIG. 1 , such an indoor air quality monitoring system 10 includes an operation server 11 , a smart sensor module 12 for measuring indoor air quality in real time, and a weather information receiving module 13 for receiving external weather information. ), and includes a ventilation or purification device 14 and a user interface 15 installed in the facility, and various information is stored and shared through the cloud 16 . In addition, the indoor air quality monitoring system 10 of this embodiment further includes an air quality level prediction module 17 for predicting a changed air quality level based on the air quality information measured by the smart sensor module 12 .

Here, the smart sensor module 12 may be composed of a plurality of sensors that detect temperature, humidity, TVOC, fine dust, carbon monoxide, carbon dioxide, etc., and further include various types of sensors for measuring information related to indoor air quality. may include

The smart sensor module 12 may be installed not only indoors (inside the facility) but also outdoors (outside the facility) as needed. In addition, the weather information receiving module 13 is linked with the Korea Meteorological Agency to collect weather information predicted in real time and in units of time from the Meteorological Agency in real time.

The ventilation or purifier 14 is various types of devices for ventilating indoors and outdoors or for purifying indoor air, and may include an air conditioner, a ventilation purifier, an air purifier, and the like.

The user interface 15 is configured to allow a user or manager to check indoor air quality information, and may include a web-accessible terminal or smart phone.

The air quality level prediction module 17 is configured to predict a new air quality level based on real-time measured air quality information, information about the facility itself, and accumulated air quality prediction information for the facility, and machine learning based on the provided information This possible artificial intelligence model is applied.

The indoor air quality monitoring system of the above configuration is installed in facilities where the vulnerable to air pollution live, such as kindergartens, daycare centers, nursing homes, nursing homes, etc. is provided to the manager or guardian, and is configured to operate a ventilation device or a purification device according to the result.

Looking at the main configuration of the indoor air quality monitoring system 10 of this embodiment for such indoor air quality prediction and control, as shown in FIG. 2 , a first information measurement unit 200, a second information collection unit ( 300 ), a weather information collecting unit 400 , a big data storage unit 500 , an air quality level prediction unit 600 , an air quality information providing unit 700 , a device driving unit 800 , and an operation server 100 .

More specifically, the first information measuring unit 200 is configured to measure real-time indoor air quality, and as shown in FIG. 3 , a temperature/humidity sensor for detecting temperature and humidity, and a human body sensor for detecting a person , TVOC (Total Volatile Organic Compounds) that detects harmful organic compounds, carbon dioxide and carbon monoxide detection sensors, and fine dust detection sensors. The first information measuring unit 100 is not limited thereto as long as it is a sensor that measures a substance that is harmful to the human body and causes air pollution, and may further include various types of detection sensors. Information such as temperature, humidity, TVOC, CO, CO2, fine dust measured by the first information measurement unit 100 is used as data for predicting changes in air quality.

The temperature/humidity sensor is configured to detect indoor temperature and humidity, and for example, may be configured as a sensor capable of detecting a temperature of -40°C to 80°C and humidity in a range of 0 to 100%. It monitors whether there is an abnormal temperature inside the facility according to the information measured by the temperature/humidity sensor.

The human body sensor detects the number (density) and activity patterns of residents living indoors, and may be configured as a heat sensor.

The TVOC detection sensor is configured to detect toxic gas generated from fire, etc., and is composed of a sensor capable of detecting toxic gas in the range of 0 to 1187 ppb.

The carbon monoxide detection sensor is to prevent gas poisoning in a colorless and odorless environment, and consists of a sensor that detects carbon monoxide in the range of 20 to 2000 ppm. The carbon dioxide detection sensor is intended to detect harmful environments in daily life, and consists of a sensor that detects carbon dioxide in the range of 400 to 8192 ppm.

The fine dust detection sensor detects dust distributed in the air, and may be configured to detect fine dust, ultra-fine dust, or ultra-fine dust separately. Depending on the particle size, dust is classified into fine dust (PM10) with a diameter of 10 μm or less, ultrafine dust with a diameter of 2.5 μm or less (PM2.5), and ultrafine dust with a diameter of 1.0 μm or less (PM1.0).

The first information measuring unit 100 as described above may include a plurality of sensors in one module, and may be installed not only indoors but also outdoors, and may be utilized as information for comparing indoor and outdoor air quality.

The second information collection unit 300 is configured to collect information on air quality change trends from big data on air quality changes. The information collected by the second information collecting unit 200 is used as information for predicting changes and levels of air quality together with the information measured by the first information measuring unit 100 . The second information collection unit 200 extracts trend information on the existing air quality change and information on the corresponding facility from the big data storage unit 500 . To this end, the big data storage unit stores and manages a large amount of information on air quality measurement and change and facility information as big data.

The weather information collection unit 400 is configured to receive weather information such as fine dust concentration and weather from an external organization. An external organization that provides meteorological information may be, for example, the Korea Meteorological Administration, and the meteorological information collection unit 400 is provided with a wired/wireless communication means for communicating with the Korea Meteorological Administration. The weather information collected by the weather information collection unit 400 includes current real-time weather information and weather information predicted in units of minutes or hours, and the collected weather information is used as information for determining whether or not the ventilation device is driven. .

The big data storage unit 500 stores and manages past and present change information of air quality according to each harmful component. The big data storage unit 500 records and manages changes in air quality in the past based on time series. The air quality information stored and managed in the big data storage unit 500 is provided to the second information collection unit 200, and real-time air quality information received from the first information measurement unit 100 is added to the past air quality change trend. to record and manage

In addition, the big data storage unit 500 stores information on facilities in which the air quality monitoring system is installed. Changes in indoor air quality are also greatly influenced by spatial characteristics such as the size or type of facilities and the level of ventilation. Therefore, the big data storage unit 500 uses the air quality and its change trend as source data, and stores the spatial size of the facility and the past ventilation time (here, the ventilation time refers to the ventilation device driving time or the purification device driving time) as structured data. and manage

As such, the information stored and managed in the big data storage unit 500 is provided according to the request of the second information collection unit 200 and is used as information for predicting the air quality level.

The air quality level prediction unit 600 is configured to predict a change in air quality based on currently measured air quality information, past air quality change trend information, and facility information, and to determine a risk level accordingly. The air quality level prediction unit 600 of this embodiment is applied with an artificial intelligence model to analyze a change trend of air quality in the past, and then predicts a change in the current air quality according to the change trend. At this time, the air quality level prediction unit 600 predicts the air quality change by additionally applying information about the facility, that is, the spatial information of the facility and the past ventilation time information. The air quality level prediction unit 600 infers in advance a time point at which air pollution reaches a critical value, that is, a time point at which the dangerous level is reached by predicting the change in air quality as described above.

The air quality information providing unit 700 provides the air quality information predicted by the air quality level prediction unit 600 to a user interface (refer to 15 of FIG. 1 ). The manager or guardian can check and monitor the air quality information in the facility in real time through the user interface, and can directly drive the ventilation device or the purification device if necessary.

The device driving unit 800 drives the ventilation device or the purification device based on the prediction result of the air quality level prediction unit 600 . That is, the air quality information providing unit 700 drives the ventilator or purifier in advance so that the air pollution does not reach the threshold when the time point at which the air pollution reaches the threshold is predicted by the air quality level prediction unit 600 . do.

In addition, in driving the ventilation device or the purification device, the device driving unit 800 ventilates by referring to the air quality information of the first information measuring unit 100 installed outside and the weather information collected by the weather information collecting unit 400 . and determine whether to drive the device or the purifier. For example, the ventilation device is driven when the outdoor air quality level is better than the indoor air quality level, and only the purification device can be driven when the outdoor air quality level is worse than the indoor air quality level due to fine dust or less than a preset standard. .

The operation server 100 includes a first information measurement unit 200 , a second information collection unit 300 , a weather information collection unit 400 , a big data storage unit 500 , an air quality level prediction unit 600 , and air quality information The operation of the providing unit 700 and the device driving unit 800 is operated and monitored.

4 is a flowchart illustrating an air quality level prediction process according to the present embodiment, FIG. 5 is a data table exemplifying information measured by the first information measurement unit according to the present embodiment, and FIG. It is a diagram illustrating structured data stored in a data storage unit, and FIG. 7 is a data table illustrating source data and structured data information of a big data storage unit according to this embodiment, and FIG. 8 is an indoor air quality level according to this embodiment. It is a data table that exemplifies the prediction results.

Referring to FIG. 4 , the air quality level prediction according to the present embodiment starts with a process in which the first information measurement unit 100 measures indoor air quality in real time ( S11 ). Indoor air quality measurement is collected by measuring indoor temperature, humidity, TVOC, fine dust (PM10, PM2.5), carbon monoxide, and carbon dioxide concentrations from the smart sensor module. The information measured by the first information measuring unit 100 may be converted into data as shown in FIG. 5 .

The measured first information is transmitted to the air quality level prediction unit 600 , and is also transmitted to the big data storage unit 500 at the same time. The big data storage unit 500 stores and manages the transmitted first information together with the air quality information in the past to check the change trend in time series (S12).

The big data storage unit 500 combines the first information with the facility information classified as shown in FIG. 6 into data as shown in FIG. 7 , and organizes and manages the data as learning data for machine learning. At this time, the big data storage unit 500 converts the transmitted information into numerical data and then non-linear processing using logarithm, non-linear processing using categorization, normalization, exception value analysis, dimensional compression using principal component analysis, sensor data loss You can build data for training by applying a processing model.

As described above, the big data storage unit 500 adds columns for space size and ventilation in addition to the hourly air quality sensor data to build learning data, thereby further improving the accuracy of air quality change prediction.

And, the air quality level prediction unit 600 receives the transmitted first air quality information and at the same time extracts big data including past air quality information and facility information from the big data storage unit 500 (S13), and the air quality trend in the past A change in air quality is predicted based on information, current air quality information, and facility information (S14), and a risk level according to the change is calculated (S15).

The predicted air quality change and the resulting risk level can be tabulated as shown in FIG. 8, and the risk level is a danger level exceeding the guide level by 100%, a warning level exceeding 90%, a caution level exceeding 80%, and other It can be classified as a safety level.

The air quality information providing unit 700 provides air quality information including the calculated risk level as a user interface (S16).

Then, the device driving unit 800 drives the ventilation device or the purification device before the air quality reaches the warning stage or the caution stage so that the air pollution does not reach the threshold (S18). At this time, the device driving unit 800 determines whether to drive the ventilation device or the purification device by referring to the external weather information collected by the weather information collecting unit 400 (S17).

As can be seen, in the indoor air quality monitoring system of this embodiment and the air quality level prediction method through it, ventilation or purification is performed in advance before the air pollution measured in real time reaches a threshold value using the past air quality change trend and facility information. In particular, it is possible to maintain a high-quality air condition for the facilities where the vulnerable class resides, such as infant facilities, disabled facilities, and nursing facilities.

As described above, exemplary embodiments of the present invention have been illustrated and described, but various modifications and other embodiments may be made by those skilled in the art. All such modifications and other embodiments are intended to be contemplated and included in the appended claims without departing from the true spirit and scope of the present invention.

100: Operation Server
200: first information measurement unit
300: second information collection unit
400: weather information collection unit
500: big data storage
600: air quality level prediction unit
700: air quality information providing unit
800: device driving unit

Claims (7)

In the air quality monitoring system installed in the facilities of the air pollution vulnerable class to monitor changes in indoor air quality,
a first information measuring unit having a sensor module for detecting a plurality of harmful substances to measure the air quality inside the facility in real time;
a big data storage unit for accumulatively storing and managing air quality information inside the facility;
a second information collection unit for collecting past air quality change information from the big data storage unit;
an air quality level prediction unit that predicts a change in air quality measured by the first information measurement unit based on the second information collected by the second information collection unit through machine learning and determines a risk level;
an air quality information providing unit that provides information on a change in air quality and a dangerous level predicted by the air quality level prediction unit to a user interface; and,
Including; including a;
An indoor air quality monitoring system that prevents air pollution inside a facility from reaching a threshold. The method of claim 1,
The big data storage unit further stores and manages facility information,
The second information collection unit further collects the facility information,
The air quality level prediction unit is configured to predict the air quality level based on the past air quality change information and the facility information, indoor air quality monitoring system. According to claim 2, wherein the facility information,
Indoor air quality monitoring system, including space size and past ventilation time information of the facility. 4. The method of claim 3,
It further includes; a meteorological information collection unit for collecting meteorological information from an external organization;
The device driving unit is configured to selectively drive a ventilation device or a purification device by comparing the air quality level predicted by the air quality level prediction unit with the weather information collected from the weather information. In the air quality level prediction method for predicting air quality changes and dangerous levels through an air quality monitoring system installed in facilities of the air pollution vulnerable group,
(a) measuring real-time indoor air quality;
(b) extracting past indoor air quality change information from the big data storage unit;
(c) predicting a change in air quality measured in step (a) and a risk level through machine learning based on the past air quality change information extracted in step (b);
(d) providing the air quality change and risk level information predicted in step (c) to a user interface; and,
(e) in response to the dangerous level of air quality predicted in step (c), before the air quality reaches a dangerous level, driving a ventilation device or a purifier; including; air quality level prediction method. 6. The method of claim 5,
The step (b) further extracts facility information including facility space size and past ventilation information of the facility from the big data storage unit,
In the step (c), the air quality level prediction method for predicting the air quality level through machine learning for the past air quality change information and the facility information. 7. The method of claim 6,
(f) receiving the air quality information measured in step (a) and ventilation information of step (e) to build data for learning; further comprising, an air quality level prediction method.

Images