• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2019년도 춘계학술대회
• /
• pp.543-545
• /
• 2019

미세먼지에 따른 행동 지침을 위해 AQI 지수가 개발되어 사용되고 있다. AQI 지수에 대한 정보는 일반인들도 쉽게 제공 받을 수 있으며, 이에 따라 AQI 지수를 기반으로 다양한 서비스가 제공되고 있다. 서비스가 제공됨에 따라 정확한 AQI 지수의 예측이 필요하다. 본 논문에서는 미세먼지의 AQI 지수를 예측하기 위해 순환 신경망을 이용하여 분류 모델의 설계를 진행한다. 설계된 모델의 평가를 위해 실제 미세먼지와 예측치의 AQI 지수를 비교한다.

• 한국대기환경학회지
• /
• 제22권4호
• /
• pp.518-528
• /
• 2006

The air quality index (AQI) is an index for reporting daily or hourly air quality to the general public. The AQI focuses on health effects that can happen within a few hours or days after breathing polluted air. Many countries have their own AQI reporting systems, and the HAEI (hourly air environment index) method is now being used in Korea. In this study, in order to compare the AQI results from different methods, we applied three methods. i.e. US AQI, Canadian AQI, and Korean HAEI, to the same air quality data-base. The data-base was constructed from 10 monitoring sites in Gyeong-buk province for the last four years since 2000. Based on the results, a critical evaluation of the Korean HAEI method was made, and a number of suggestions and recommendations were presented to improve the AQI reporting system in Korea.

• Asian Journal of Atmospheric Environment
• /
• 제9권2호
• /
• pp.101-113
• /
• 2015

Air quality index (AQI) or air pollution index (API) is commonly used to report the level of severity of air pollution to public. A number of methods were developed in the past by various researchers/environmental agencies for determination of AQI or API but there is no universally accepted method exists, which is appropriate for all situations. Different method uses different aggregation function in calculating AQI or API and also considers different types and numbers of pollutants. The intended uses of AQI or API are to identify the poor air quality zones and public reporting for severity of exposure of poor air quality. Most of the AQI or API indices can be broadly classify as single pollutant index or multi-pollutant index with different aggregation method. Every indexing method has its own characteristic strengths and weaknesses that affect its suitability for particular applications. This paper attempt to present a review of all the major air quality indices developed worldwide.

• 국제초고층학회논문집
• /
• 제6권1호
• /
• pp.101-111
• /
• 2017

The objective of this study is to determine the effectiveness of a modified air-side economizer in improving indoor air quality (IAQ). An air-side economizer, which uses all outdoor air for cooling, affects the building's IAQ depending on the outside air quality and can significantly affect the occupants' health, leading to respiratory and heart disease. The Air Quality Index (AQI), developed by the US Environmental Protection Agency (US EPA), measures air contaminants that adversely affect human beings: PM10, PM2.5, SO2, NO2, O3, and CO. In this study, AQI is applied as a control for the operation of an air-side economizer. The simulation is analyzed, comparing the results between the differential enthalpy economizer and AQI-modified economizer. The results confirm that an AQI-modified economizer has a positive effect on IAQ. Compared to the operating differential enthalpy economizer, energy increase in an operating AQI-modified economizer is 0.65% in Shanghai and 0.8% in Seoul.

• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2019년도 춘계학술대회
• /
• pp.540-542
• /
• 2019

미세먼지로 인한 대기오염 및 인체 영향에 대한 많은 발표로 인해 미세먼지 예보는 많은 대중의 관심을 받고 있다. 이로 인해 통계 모델링 기법과 함께 기계학습 기법을 사용하여 미세먼지 예보 정확도를 올리기 위한 다양한 노력이 수행되고 있다. 본 논문에서는 미세먼지 예측을 위해 다층 퍼셉트론 신경망을 활용한 미세먼지 AQI 지수 예측을 수행한다. 이를 위해 다수의 연구에서 공통적으로 사용된 기상 인자와 미세먼지 농도값을 이용하여 예측 모델을 설계하고 4단계의 미세먼지 AQI 예측 정확도를 비교한다.

• Advances in environmental research
• /
• 제4권2호
• /
• pp.83-104
• /
• 2015

Air Quality Index (AQI) is a pointer to broadcast short term air quality. This paper presents one day ahead AQI forecasting on seasonal basis for three major cities in Maharashtra State, India by using Artificial Neural Networks (ANN) and Genetic Programming (GP). The meteorological observations & previous AQI from 2005-2008 are used to predict next day's AQI. It was observed that GP captures the phenomenon better than ANN and could also follow the peak values better than ANN. The overall performance of GP seems better as compared to ANN. Stochastic nature of the input parameters and the possibility of auto-correlation might have introduced time lag and subsequent errors in predictions. Spectral Analysis (SA) was used for characterization of the error introduced. Correlational dependency (serial dependency) was calculated for all 24 models prepared on seasonal basis. Particular lags (k) in all the models were removed by differencing the series, that is converting each i'th element of the series into its difference from the (i-k)"th element. New time series is generated for all seasonal models in synchronization with the original time line & evaluated using ANN and GP. The statistical analysis and comparison of GP and ANN models has been done. We have proposed a promising approach of use of GP coupled with SA for real time prediction of seasonal multicity AQI.

• 한국조경학회지
• /
• 제49권5호
• /
• pp.1-11
• /
• 2021

본 연구는 도시의 표면을 구성하는 물리적 공간요소 가운데 대기질에 영향을 미치는 주요 요인을 찾아 환경개선을 위한 단서를 제공하는 것이 목적이다. 중국의 산업도시인 창춘의 9개 측정소에서 2018년 1월 1일부터 2019년 12월 31일까지의 AQI 농도 자료를 수집하였다. 측정소를 중심으로 반경 300m 내의 지역을 구성하는 도시의 물리적 시설에 대한 유형과 분포 특징을 분석하였다. 측정소를 3개 그룹으로 나눠 계절별로 미세먼지 농도 차이를 분석한 결과, 봄과 겨울에 AQI 농도가 가장 높고, 다음으로 여름, 가장 낮은 계절은 가을이었다. 봄의 AQI 농도가 가장 높은 곳은 F(93.00)·D(91.10)·I(89.20), 여름 농도가 가장 높은 곳은 D(69.05)·A(67.89)·B(84.44), 가을 농도가 가장 높은 곳은 I(62.80)·G(60.84)·D(53.27), 겨울은 I(95.82)·H(95.60)·F(94.04)이었다. SPSS를 이용한 계열분석을 통해 직경 600m인 공간내의 대기지수는 임야, 초지, 나지, 수공간, 수고, 건축면적(평균치), 건물 체적(평균치)과 상관성이 있는 것으로 나타났다. 오염이 심한 봄과 겨울의 수치자료를 통계분석한 결과, 임야면적(43,637m², 15.44%)과 수면적(18,736m², 6.63%)이 큰 비율을 차지하고, 건축평균면적(448m², 0.17%)과 건축평균부피(10,201m³)가 가장 적은 그룹 1(A, B, C)구역의 오염 농도가 가장 낮았다. 반대로 그룹 2(D, E, F)구역은 AQI 농도가 가장 높은 구역으로 임야(1,917m², 0.68%)와 수면적(0m², 0%)이 적거나 없고 건축평균면적(1,056m², 0.37%)과 건축평균부피(17,470m³)가 가장 높았다. AQI 농도가 가장 높은 지역의 특징은 가로수의 수고가 12m 이상인 경우가 다수 확인되었고, 나지면적의 비율이 적을수록 오염도가 낮은 것으로 나타났다. 동일한 방법으로 창춘의 9개 공간의 특징을 분석한 결과, 도시 공간의 물리적 특성에 따른 대기질은 위의 요인들과 밀접한 관련이 있음을 알 수 있었다.

• Environmental Engineering Research
• /
• 제23권1호
• /
• pp.21-27
• /
• 2018

Air pollution index (API) is used in Mainland China and includes only $SO_2$, $NO_2$ and $PM_{10}$. In 2016, air quality index (AQI) replaced API. AQI contains three more air pollutants (CO, $O_3$ and $PM_{2.5}$). Both the indices emphasize on the effect of a single pollutant, whereas the contributions of all other pollutants are ignored. Therefore, in the present work, a novel air quality index (NAQI), which emphasizes on all air pollutants, has been introduced for the first time. The results showed that there were 19 d (5.2%) in API, 28 d (7.7%) in AQI and 183 d (50.1%) in NAQI when the indices were more than 100. In API, $PM_{10}$ and $SO_2$ were regarded as the primary pollutants, whereas all five air pollutants in AQI were regarded as primary. Furthermore, four air pollutants (other than the CO) in NAQI were regarded as primary pollutants. $PM_{10}$, as being the primary pollutant, contributed greatly in these air quality indices, and accounted for 51.2% (API), 37.0% (AQI) and 52.6% (NAQI). The results also showed that particulate matter pollution was significantly high in Luzhou, where stricter pollution control measures should be implemented.

• Communications for Statistical Applications and Methods
• /
• 제30권2호
• /
• pp.119-133
• /
• 2023

With the rapid growth of the economy and fossil fuel consumption, the concentration of air pollutants has increased significantly and the air pollution problem is no longer limited to small areas. We conduct statistical analysis with the actual data related to air quality that covers the entire of South Korea using R and Python. Some factors such as SO₂, CO, O₃, NO₂, PM₁₀, precipitation, wind speed, wind direction, vapor pressure, local pressure, sea level pressure, temperature, humidity, and others are used as covariates. The main goal of this paper is to predict air quality index (AQI) spatio-temporal data. The observations of spatio-temporal big datasets like AQI data are correlated both spatially and temporally, and computation of the prediction or forecasting with dependence structure is often infeasible. As such, the likelihood function based on the spatio-temporal model may be complicated and some special modelings are useful for statistically reliable predictions. In this paper, we propose several methods for this big spatio-temporal AQI data. First, random effects with spatio-temporal basis functions model, a classical statistical analysis, is proposed. Next, neural networks model, a deep learning method based on artificial neural networks, is applied. Finally, random forest model, a machine learning method that is closer to computational science, will be introduced. Then we compare the forecasting performance of each other in terms of predictive diagnostics. As a result of the analysis, all three methods predicted the normal level of PM_2.5 well, but the performance seems to be poor at the extreme value.

• 한국정보통신학회논문지
• /
• 제24권1호
• /
• pp.8-14
• /
• 2020

미세먼지의 인체 영향이 밝혀지며 예보정확도 개선에 대한 요구가 증가하고 있다. 이에 기계 학습 기법을 도입하여 예측 정확성을 높이려는 노력이 수행되고 있으나, 저농도 발생 비율이 매우 큰 미세먼지 데이터로 인해 전체 예측 성능이 떨어지는 문제가 있다. 본 논문에서는 PM10 미세먼지 예보 정확도 향상을 위해 농도별 분리 예측 모델을 제안한다. 이를 위해 천안 지역의 기상 및 대기오염 인자를 활용하여 저, 고농도별 예측 모델을 설계하고 전 영역 예측 모델과의 성능 비교를 수행하였다. RMSE, MAPE, 상관계수 및 AQI 정확도를 통한 성능 비교 결과, 전체 기준에서 예측 성능이 향상됨을 확인하였으며, AQI 고농도 예측 성능의 경우 20.62%의 성능 향상이 나타났음을 확인하였다.