Comparative Analysis of Daily and Weekly Heavy Rain Prediction Using LSTM and Cloud Data

Monita, Vivi and Raniprima, Sevirda and Cahyadi, Nanang (2024) Comparative Analysis of Daily and Weekly Heavy Rain Prediction Using LSTM and Cloud Data. Jurnal Ilmiah Teknik Elektro Komputer dan Informatika (JITEKI), 10 (4). pp. 833-842.

Text 13-Comparative Analysis of Daily and Weekly Heavy Rain Prediction Using LSTM and Cloud Data.pdf Download (928kB)

Abstract

Indonesia's distinct geographic and climatic features make forecasting the weather there tricky. Due to its location at the equator and between two enormous oceans, the nation endures erratic weather patterns. Despite technical developments, the Meteorology, Climatology, and Geophysics Agency (BMKG) require assistance with precise forecasting. This research seeks to increase prediction accuracy using the Long Short-Term Memory (LSTM) algorithm, a deep learning technique appropriate for time series data processing. The research focuses on cloud data sets to improve the prediction of heavy rain. The potential of LSTM in weather forecasting has been demonstrated in earlier research, focusing on identifying rain at particular intervals. This research compares daily and weekly heavy rain prediction models using Python. Results reveal that the weekly model outperforms the daily model, achieving 85% accuracy compared to 80%. These findings highlight the effectiveness of LSTM in addressing the limitations of existing methods, offering a foundation for more reliable weather forecasting tailored to Indonesia’s conditions.

Item Type:	Artikel Umum
Subjects:	T Technology > TK Electrical engineering. Electronics Nuclear engineering
Divisi / Prodi:	Faculty of Industrial Technology (Fakultas Teknologi Industri) > S1-Electrical Engineering (S1-Teknik Elektro)
Depositing User:	M.Eng. Alfian Ma'arif
Date Deposited:	21 Feb 2025 02:20
Last Modified:	21 Feb 2025 02:20
URI:	http://eprints.uad.ac.id/id/eprint/82002

Actions (login required)

View Item