Current Result Document :
| 한글제목(Korean Title) |
LSTM모델을 이용한 지하수위 예측에서의 기상학적 요인의 영향에 대한 분석 |
| 영문제목(English Title) |
Analysis of Meteorological factors on Groundwater Level variations using Long Short-Term Memory Network |
| 저자(Author) |
Adnan Haider
Pilsun Yoon
Heedong Park
Kyoungson Jhang
어드난 하이더
윤필선
박희동
장경선
|
| 원문수록처(Citation) |
VOL 49 NO. 02 PP. 0175 ~ 0177 (2022. 12) |
한글내용
(Korean Abstract) |
|
영문내용
(English Abstract) |
To accurately simulate the groundwater level (GWL), it is of great importance to study the impact of the external meteorological factors (i.e., temperature (T), precipitation (P)) that cause variations in the groundwater level both in the short and long term. In this study, we developed two LSTM models with different set of input features and the results show that if external factors are used as input parameters along with past groundwater level time series, the model can learn to simulate short-term fluctuations more accurately. However, in some cases model trained with single input outperformed multi-input model. Furthermore, the time lag between GWL and external factors was not calibrated in data preprocessing, we believe by adding time lag will further improve the simulations of multi-input variables model. |
| 키워드(Keyword) |
|
| 파일첨부 |
PDF 다운로드
|
