Research for Development of a Daejeon Fine Dust Prediction Model through Weather Data and Air Pollutants
Tae-Hyung Kim1, Jae-Ou Lee2
1Tae-Hyung Kim, Department of Fire and Disaster Prevention, Daejeon University, Daejeon City, Korea, East Asian.
2Jae-Ou Lee, Department of Fire and Disaster Prevention, Daejeon University, Daejeon City, Korea, East Asian.
Manuscript received on 10 June 2019 | Revised Manuscript received on 17 June 2019 | Manuscript Published on 22 June 2019 | PP: 737-743 | Volume-8 Issue-8S2 June 2019 | Retrieval Number: H11230688S219/19©BEIESP
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open-access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Fine dust is a first-tier carcinogen that is adversely affecting animals, plants, and people. Since 2005, the Ministry of Environment has been implementing fine dust forecasts to manage such fine dust. The forecast study conducted research on the causes of fine dust and how to cope with it. Methods/Statistical analysis: In this study, a model for predicting fine dust was developed through phase of multiple regression analysis. Independent variable can largely sort the forecast data and the air pollutants, and the dependent variable is fine dust (PM10 ). It consists of a total of fifty variables and data has been collected on a monthly basis from 2006 to 2017. Findings: The result of regression from the model with the highest explanatory power shows that at the magnitude of the definition, a large total evaporation loss was the highest with .492 followed by CO with .264. Partial effect was 1.0m, average ground temperature was -.566, and highest temperature was -.325 in order. Therefore, fine dust (PM10 ) increases when the large total evaporation loss and CO concentration increases. On the other hand, fine dust decreases when the average ground temperature and the highest temperature decreases. As a result, fine dust is influenced by diverse factors, and their effects show both increase and decrease. Improvements/Applications: Through this research, to identify factors that influence fine dust and use those in areas to manage fine dust in the future. In the final fine dust prediction model, the 1.0m average underground temperature and the maximum temperature were negatively affected, while the carbon monoxide and the large layer emissions were positively affected.
Keywords: Fine Dust, Phases of Regression Analysis, Weather Data, Air Pollution, Multiple Linear Regression Model.
Scope of the Article: Analysis of Algorithms and Computational Complexity