Weather Forecasting Using Artificial Neural Network

Abstract

Weather forecasting is a blessing of modern technology. It enables us to understand the nature of the atmosphere. Precise weather forecasting is one of the greatest challenges in the modern world. Unlike traditional methods, modern weather forecasting involves a combination of computer models, observation (by use of balloons and satellites) and patterns recognition along with various trends. Forecasting can be made accurately and precisely by the proper application of these methods. For forecasting various kinds of computer methods are used and these methods are related to various complex formulas. Researchers have done many things to establish a relationship of recent (input) data and target data which is linear. But practically the relationship is nonlinear. After establishing the nonlinearity, many models have been made to get future weather data. As the weather data is nonlinear, Artificial Neural Network (ANN) has become an effective way of predicting weather data precisely and accurately. Neural Network is a system that can be trained with certain input and output. It creates its own structure based upon how it is trained. In this paper we predicted weather data for a particular month of a season and compared the result for different functions and training method of ANN.