멤버십 함수와 DNN을 이용한 PM10 예보 성능의 향상

유숙현,전영태,권희용 한국멀티미디어학회 2019 멀티미디어학회논문지 Vol.22 No.9

In this study, we developed a PM10 forecasting model using DNN and Membership Function, and improved the forecasting performance. The model predicts the PM10 concentrations of the next 3 days in the Seoul area by using the weather and air quality observation data and forecast data. The best model(RM14)’s accuracy (82%, 76%, 69%) and false alarm rate(FAR:14%,33%,44%) are good. Probability of detection (POD: 79%, 50%, 53%), however, are not good performance. These are due to the lack of training data for high concentration PM10 compared to low concentration. In addition, the model dose not reflect seasonal factors closely related to the generation of high concentration PM10. To improve this, we propose Julian date membership function as inputs of the PM10 forecasting model. The function express a given date in 12 factors to reflect seasonal characteristics closely related to high concentration PM10. As a result, the accuracy (79%, 70%, 66%) and FAR (24%, 48%, 46%) are slightly reduced in performance, but the POD (79%, 75%, 71%) are up to 25% improved compared with those of the RM14 model. Hence, this shows that the proposed Julian forecast model is effective for high concentration PM10 forecasts.

Update of the year 2019 modeling emission inventory in China

김서연,김진석,허혜정,장명도,이재범,홍성철,김옥길,우정헌 한국대기환경학회 2023 Asian Journal of Atmospheric Environment (AJAE) Vol.17 No.1

Using updated emission inventories can enhance the accuracy of air quality forecast models. Given China’s rapid economic growth and Korea’s geographical and meteorological position on the windward side of China, updating China’s emission inventory has become particularly crucial for Korea’s air quality modeling. This study aimed to develop an updated version of China’s Emission Inventory in Comprehensive Regional Emissions for Atmospheric Transport Experiments version 3 for the base year of 2019 (CREATEv3 (YR 2019)). To achieve this goal, we utilized the Chinese emission inventory of CREATEv3 for the base year of 2015 (CREATEv3 (YR 2015)) as a framework to incorporate the latest Chinese emission data from the Multi-resolution Emission Inventory Model for Climate and Air Pollution Research for the base year of 2019 (MEIC COVID-19 (YR 2019)) and update the inventory. The updated China’s annual emissions are now reflected in CREATEv3 (YR 2019), and the amounts are as follows: 132 Tg for CO, 21 Tg for NOx, 8 Tg for SO2, 7 Tg for PM2.5, 9 Tg for NH3, and 28 Tg for volatile organic compound (VOC). By comparing previous Chinese emission inventories with the updated inventory developed in this study, it was found that SO2, NOx, VOC, and NH3 emissions were decreased. Therefore, using the updated inventory seemingly reduces the impact of China’s fine dust on Korea. By comparing emissions by pollutant and region in China using CREATEv3 (YR 2019), it was found that regions with high emissions of targeted pollutants strongly correlated with major industries operating in those areas. This study is expected to provide insights into China’s emission changes in 2019 and support air quality forecasting.

국가 대기질 예보 시스템의 모델링(기상 및 대기질) 계산속도 향상을 위한 전산환경 최적화 방안

명지수 ( Jisu Myoung ),김태희 ( Taehee Kim ),이용희 ( Yonghee Lee ),서인석 ( Insuk Suh ),장임석 ( Limsuk Jang ) 한국환경과학회 2018 한국환경과학회지 Vol.27 No.8

In this study, to investigate an optimal configuration method for the modeling system, we performed an optimization experiment by controlling the types of compilers and libraries, and the number of CPU cores because it was important to provide reliable model data very quickly for the national air quality forecast. We were made up the optimization experiment of twelve according to compilers (PGI and Intel), MPIs (mvapich-2.0, mvapich-2.2, and mpich-3.2) and NetCDF (NetCDF-3.6.3 and NetCDF-4.1.3) and performed wall clock time measurement for the WRF and CMAQ models based on the built computing resources. In the result of the experiment according to the compiler and library type, the performance of the WRF (30 min 30 s) and CMAQ (47 min 22 s) was best when the combination of Intel complier, mavapich-2.0, and NetCDF-3.6.3 was applied. Additionally, in a result of optimization by the number of CPU cores, the WRF model was best performed with 140 cores (five calculation servers), and the CMAQ model with 120 cores ( five calculation servers). While the WRF model demonstrated obvious differences depending on the number of CPU cores rather than the types of compilers and libraries, CMAQ model demonstrated the biggest differences on the combination of compilers and libraries.

A Study on Particulate Matter Forecasting Improvement by using Asian Dust Emissions in East Asia

Daeryun Choi,Huiyoung Yun,Limseok Chang,Jaebum Lee,Younghee Lee,Jisu Myoung,Taehee Kim,Younseo Koo 한국도시환경학회 2018 한국도시환경학회지 Vol.18 No.4

Air quality forecasting system with Asian dust emissions was developed in East Asia, and PM10 forecasting performance of chemical transport model with Asian dust emissions was validated and evaluated. The chemical transport model (CTM) with Asian dust emission was found to supplement PM10 concentrations that had been under-estimated in China regions and improved statistics for performance of CTM, although the model were overestimated during some periods in China. In Korea, the prediction model adequately simulated inflow of Asian dust events on February 22~24 and March 16~17, but the model is found to be overestimated during no Asian dust event periods on April. However, the model supplemented PM10 concentrations, which was underestimated in most regions in Korea and the statistics for performance of the models were improved. The PM10 forecasting performance of air quality forecasting model with Asian dust emissions tends to improve POD (Probability of Detection) compared to basic model without Asian dust emissions, but A (Accuracy) has shown similar or decreased, and FAR (False Alarms) have increased during 2017.Therefore, the developed air quality forecasting model with Asian dust emission was not proposed as a representative PM10 forecast model in South Korea.

Exploiting Neural Network for Temporal Multi-variate Air Quality and Pollutant Prediction

Muneeb A. Khan,김현철,박희민 한국멀티미디어학회 2022 멀티미디어학회논문지 Vol.25 No.2

In recent years, the air pollution and Air Quality Index (AQI) has been a pivotal point for researchers due to its effect on human health. Various research has been done in predicting the AQI but most of these studies, either lack dense temporal data or cover one or two air pollutant elements. In this paper, a hybrid Convolutional Neural approach integrated with recurrent neural network architecture (CNN- LSTM), is presented to find air pollution inference using a multivariate air pollutant elements dataset. The aim of this research is to design a robust and real-time air pollutant forecasting system by exploiting a neural network. The proposed approach is implemented on a 24-month dataset from Seoul, Republic of Korea. The predicted results are cross-validated with the real dataset and compared with the state-of-the-art techniques to evaluate its robustness and performance. The proposed model outperforms SVM, SVM-Polynomial, ANN, and RF models with 60.17%, 68.99%, 14.6%, and 6.29%, respectively. The model performs SVM and SVM-Polynomial in predicting O3 by 78.04% and 83.79%, respectively. Overall performance of the model is measured in terms of Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE) and the Root Mean Square Error (RMSE).

황사배출량을 적용한 동아시아 미세먼지 예보 개선 연구

최대련,윤희영,장임석,이재범,이용희,명지수,김태희,구윤서 한국도시환경학회 2018 한국도시환경학회지 Vol.18 No.4

Air quality forecasting system with Asian dust emissions was developed in East Asia, and PM10 forecasting performance of chemical transport model with Asian dust emissions was validated and evaluated. The chemical transport model (CTM) with Asian dust emission was found to supplement PM10 concentrations that had been under-estimated in China regions and improved statistics for performance of CTM, although the model were overestimated during some periods in China. In Korea, the prediction model adequately simulated inflow of Asian dust events on February 22~24 and March 16~17, but the model is found to be overestimated during no Asian dust event periods on April. However, the model supplemented PM10 concentrations, which was underestimated in most regions in Korea and the statistics for performance of the models were improved. The PM10 forecasting performance of air quality forecasting model with Asian dust emissions tends to improve POD (Probability of Detection) compared to basic model without Asian dust emissions, but A (Accuracy) has shown similar or decreased, and FAR (False Alarms) have increased during 2017.Therefore, the developed air quality forecasting model with Asian dust emission was not proposed as a representative PM10 forecast model in South Korea. 동아시아지역을 대상으로 황사배출량 산정 모듈 및 이를 적용한 예보시스템을 개발하였고, 개발된 모형의 화학수송모델링 정합도 및 실시간 예보 운영 평가를 진행하였다. 2015년 화학수송모델링 정합도 평가 결과, 중국 지역에서는 황사배출량을 적용한 예보 모형이 과대평가하는 기간이 있으나 대부분 지역에서 저평가 되었던 PM10을 보완하고, 통계수치가 개선되는 것을 확인할 수 있었다. 한국 지역에서는 황사 발생일인 2월 22일~24일, 3월 16일~17일(서울지역대상)에는황사의 유입을 적절히 모사하였으나 황사가 관측되지 않은 4월에는 황사를 적용한 예보모델이 과대평가하는 것을 확인할 수 있었다. 그러나 황사를 적용한 예보모형은 한반도 대부분 지역에서 저평가 되었던 PM10을 보완하고, 통계수치가개선되는 것을 확인할 수 있었다. 2017년 예보 성능 평가 결과, 황사배출량을 적용한 예보모델은 기존 모델과 비교하였을 때, POD는 대부분 개선되지만, A는 유사 또는 감소, FAR는 대부분 증가하는 경향이 나타났다. 황사배출량을 적용한예보모형은 동아시아 지역에 저평가 하고 있는 PM10을 보완하는 장점이 있지만, 황사배출량 산정의 불확실성 등이 내제되어 모델이 측정값을 과대모의하여 오경보율이 높다. 따라서 한반도 지역에 대표 대기질 예보모형으로 사용하기는 부적절하다고 판단된다. 그러나 황사 기간에는 황사배출량 모델의 모사성능은 우수하였으므로, 황사가 발생하는 기간에는 기존 모델과 융합하여 예보관이 예보하는 것이 필요하다고 사료된다.

Improving PM<SUB>10</SUB> and PM<SUB>2.5</SUB> Forecasts in Korea using the EnKF Aerosol Data Assimilation

Seunghee Lee,Myong-In Lee,Ganghan Kim 한국기상학회 2021 한국기상학회 학술대회 논문집 Vol.2021 No.10

The air quality of South Korea is greatly influenced by long-range transport from China under westerlies as well as local sources. Therefore, accurate air quality prediction of Korea are challenges associated with numerical model forecasts, such as considerable uncertainties of the initial conditions (ICs), the physical and chemical aspects of the parameterized model, and the various emissions. The aerosol data assimilation (DA) system could significantly improve aerosol forecasting skills by reducing these uncertainties. This study developed the aerosol data assimilation and forecast system using Weather Research and Forecasting model with Chemistry (WRF-Chem) model and Ensemble Kalman Filter (EnKF) method. Unlike the variational assimilation method, the advantages of EnKF are flow-dependent background error covariance which is important in a fast-developing air quality system. The optimization of EnKF is very important since EnKF has a sampling problem due to limited ensemble number. This study optimized the EnKF system by alleviating the sampling problem through a multiphysics approach, emission perturbation, inflation, and covariance localization. The assimilated observations are aerosol optical depth (AOD) retrieved from the Geostationary Ocean Color Imager (GOCI) and the Moderate Resolution Imaging Spectroradiometer (MODIS), surface PM10 and surface PM2.5. of China and South Korea. The quality of the DA analysis fields and the prediction skill were evaluated quantitatively over South Korea. The simulations with DA initialization showed significant improvements in the PM10 and PM2.5 forecast skills.

Assimilation of AOD retrievals to improve air quality forecasts using Machine Learning and Data Assimilation

Seunghee Lee,Myong-In Lee,Seohui Park,Jungho Im 한국기상학회 2020 한국기상학회 학술대회 논문집 Vol.2020 No.10

Forecast skill dependence on the initial time in the WRF-Chem air quality forecasts

Ganghan Kim(김강한),Myong-In Lee,Seunghee Lee,Chang-Keun Song 한국기상학회 2019 한국기상학회 학술대회 논문집 Vol.2019 No.05

Improving air quality forecasts in Korea using the 3D-Var aerosol data assimilation

Seunghee Lee,Myong-In Lee,Chang-Keun Song,Ganghan Kim 한국기상학회 2019 한국기상학회 학술대회 논문집 Vol.2019 No.10