Comparative Study of Neural Network Models for Prediction of Chaotic System

Abstract

Research into the ability to forecast the internal dynamics of chaotic systems is vital. The chaotic character of climatic systems is a major barrier to accurate climate prediction. Numerous studies have been conducted so far on the topic of numerical simulation approaches for the simulation and prediction of chaotic systems. Chaotic systems are difficult to anticipate with numerical simulation approaches due to issues like sensitivity to beginning values, error accumulation, and inappropriate parameterization of physical processes. Here, we looked into the architectures of Neural Networks. The present literature study has provided confidence in the efficacy of artificial neural networks as a powerful tool for predicting internal dynamics climatological data. Research demonstrates that ANNs have the potential for prediction of climatological analysis since they are ideally adapted to situations requiring complex nonlinear interactions.