Using multiple linear regression model to estimate thunderstorm activity

International Conference on Information Technology and Digital Applications

计算机科学

Suparta, W.^1 ; Putro, W.S.^1

Space Science Centre (ANGKASA), Institute of Climate Change, Universiti Kebangsaan Malaysia, Bangi, Selangor Darul Ehsan

43600, Malaysia^1

关键词: Meteorological data;    Monsoon season;    Multiple linear regression models;    Non-linear model;    Precipitable water vapor;    Thunderstorm activity;   

Others  :  https://iopscience.iop.org/article/10.1088/1757-899X/185/1/012023/pdf
DOI  :  10.1088/1757-899X/185/1/012023

学科分类：计算机科学（综合）

来源: IOP

【 摘 要 】

This paper is aimed to develop a numerical model with the use of a nonlinear model to estimate the thunderstorm activity. Meteorological data such as Pressure (P), Temperature (T), Relative Humidity (H), cloud (C), Precipitable Water Vapor (PWV), and precipitation on a daily basis were used in the proposed method. The model was constructed with six configurations of input and one target output. The output tested in this work is the thunderstorm event when one-year data is used. Results showed that the model works well in estimating thunderstorm activities with the maximum epoch reaching 1000 iterations and the percent error was found below 50%. The model also found that the thunderstorm activities in May and October are detected higher than the other months due to the inter-monsoon season.