Publication

Publication Type : Journal Article

Publisher : AIP Publishing

Source : AIP Conference Proceedings

Url : https://doi.org/10.1063/5.0153980

Campus : Kochi

School : School of Physical Sciences

Department : Mathematics

Year : 2023

Metaheuristic optimization has gained its popularity over the last decades due to its application to solve real world problems. In this study, a new metaheuristic model, multiple linear particle swarm optimization (MLPSO) is developed to predict the rainfall of India for the monsoon season. For this purpose, daily rainfall data of India for 73 years is collected from Indian meteorological department. The dynamical parameters such as sea surface temperature, specific humidity, meridional and zonal wind, geopotential height and pressure, etc. are selected that affect rainfall over India. To evaluate the efficiency of the proposed model, various statistical measures are considered and compared with traditional multiple linear regression  model. It is found that the proposed model has better accuracy depicting root mean square error of 0.981, correlation coefficient 0.376, mean absolute error 0.863 and efficiency coefficient -0.005.

Cite this Research Publication : Chalissery Mincy Thomas, Archana Nair, A new metaheuristic model to predict rainfall by using dynamical parameters, AIP Conference Proceedings, AIP Publishing, 2023, https://doi.org/10.1063/5.0153980