Addressing Challenges and Demands of Intelligent Seasonal Rainfall Forecasting using Artificial Intelligence Approach

Abstract

Precipitation forecasting is one of the most important and stimulating tasks in the modern world. Climate and precipitation are generally highly non-linear and complex phenomena that require advanced computer modeling and simulation for their accurate prediction. An artificial neural network (ANN) can be used to predict the behavior of such nonlinear systems. RNA has been successfully used by most researchers in this field over the past 25 years. This article examines the literature available on some of the methods used by various researchers to use the AN for precipitation forecasting. Weather forecasts have become an important research field in recent decades. In most cases, the researcher sought to establish a linear relationship between the meteorological data entered and the corresponding target data. However, with the discovery of non-linearity in the nature of meteorological data, the emphasis was placed on the non-linear forecast of meteorological data. The article examines the applicability of the ANN approach by developing efficient and reliable non-linear predictive models for meteorological analysis. Furthermore, they compare and evaluate the performance of the models developed with different transfer functions, hidden layers and neurons to predict the maximum temperature 365 days a year.