2020 Fiscal Year Final Research Report
Development of determination and prediction method of long-range transport event of atmospheric aerosol particles using machine learning algorithm
| Project/Area Number |
18K04407
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Multi-year Fund |
|---|
| Section | 一般 |
|---|
| Review Section |
Basic Section 22060:Environmental systems for civil engineering-related
|
|---|
| Research Institution | University of Yamanashi |
|---|
Principal Investigator |
|
|---|
| Project Period (FY) |
2018-04-01 – 2021-03-31
|
| Keywords | 越境輸送微粒子 / 偏光光散乱式粒子計測器 / ディープラーニング / 鉱物粒子 / 大気汚染粒子 / 観測ネットワーク / 畳み込みニューラルネットワーク |
|---|
| Outline of Final Research Achievements |
In this study, we continued to add new observation site at the foot of Mt. FUji to the polarization OPC observation network to determine and predict events of long-range transported particles. We have improved the method for estimating the concentration by composition, and the accuracy of the estimation of the particle size and mass concentration of mineral particles and deliquescent particles, which are often included in anthropogenic secondary particles, has been greatly improved. A model that can determine long-range transport events has been developed using deep learning. As for the prediction, we have not been able to obtain sufficient results at present, so we will continue to improve the model in the future.
|
| Free Research Field |
大気科学
|
| Academic Significance and Societal Importance of the Research Achievements |
環境基準に微小粒子状物質が加えられたことや報道により北京での高濃度の様子が伝えられることにより，広く社会に微粒子の健康影響が意識されるようになっている．このような状況で微粒子の関する観測ネットワークを維持し，そのデータを公開するだけではなく濃度予測にも活用を試みたことが社会的な意義が高い．また，特許を取得しているオリジナルの測定器による観測ネットワークは世界で唯一であり，学術的にも貴重なデータである．
|
