Scientific Analysis of Various Computational Intelligence Methods used for Weather Forecasting

Abstract

Numerous studies in the field of weather forecasting have been conducted since technology started to develop in order to better understand how to manage weather by employing the appropriate kind of forecasting. This work is an evaluation report of facts and figures from the literature on weather forecasting incorporate with machine learning(ML), and deep learning(DL) models. Meteorologists, scientists, and researchers have created a wide range of designs, models, simulation systems, and prototypes to increase prediction accuracy. The first portion examines the literature on previous weather forecasting work, the application of numerous ML and DL models for weather forecasting along with the associated challenges. The second Section is all about the analysis  description and drawbacks of current DL weather forecasting models. Several flaws were discovered following the study of prior models. The most common concerns are that running several equations simultaneously which are non-linear in nature requires a significant amount of computer resources and takes a long time to process. computer using Data-driven modeling techniques can be used to reduce the complexity of earlier models. ML and DL, in particular, can more accurately reflect a physical process's nonlinear or intricate underlying features.