Thaha, M. Mohammed and Mohd Shaharanee, Izwan Nizal and Murugan, V. P. and Kumar Kusumanchi, T. P. S. and Lokeswa1 Recldy, D. V. (2024) Machine learning methods based on storm surge disaster loss in computing applications. International Journal of Intelligent Systems and Applications in Engineering, 12 (5s). pp. 379-389. ISSN 21476799

Preview

PDF - Published Version Available under License ["licenses_description_cc4_by_sa" not defined]. Download (2MB) | Preview

Abstract

Storm surge, which impacts the entire coastline region, is China's most serious marine calamity. Storm surge disaster loss (SSDL) estimation is important for decision-making, sustainability, and disaster prevention. For early warning systems, disaster management, and disaster evaluation, an accurate storm surge water level forecast is essential. Comparing machine learning techniques to numerical simulation techniques, the former is more straightforward and efficient. Still, point predictions are the main focus of the majority of current machine learning-based research on storm surge prediction. In this paper, we explore the feasibility of employing the ConvLSTM model for spatial water level prediction. The ConvLSTM-based methodology is simpler, faster, and more accurate in predicting water levels without the need for boundary conditions or topography than standard numerical simulation methods. In addition, we take worst-case situations into account by employing the random forest model to anticipate the highest possible water increase. Based on our findings, the random forest model may prove to be a useful instrument in determining the highest water increase value linked to typhoon storm surges, which can help with efficient emergency reactions during natural disasters

Item Type:	Article
Uncontrolled Keywords:	Storm surge, ConvLSTM, natural disasters, efficient, machine learning, water level
Subjects:	T Technology > TD Environmental technology. Sanitary engineering
Divisions:	School of Quantitative Sciences
Depositing User:	Mr. Izhar Hj. Hashim
Date Deposited:	05 Nov 2026 09:35
Last Modified:	05 Nov 2026 09:35
URI:	https://repo.uum.edu.my/id/eprint/34814

Actions (login required)

View Item