Prediction of temperature for various pressure levels using ANN and multiple linear regression techniques: A case study

Abstract

In this paper, the estimation capacities of Artificial Neural Network (ANN) and Multiple Linear Regression (MLR) are examined to forecast temperature at two pressure level 400 hPa and 980 hPa approximately over Delhi Safdarjung Airport Station. Different meteorological parameters such as temperature, relative humidity, dew point temperature, mixing ratio, potential temperature has been taken as an input for ANN and MLR models. Statistical bases give us forecasting of temperature by using above mentioned parameters. ANN gives us better results in forecasting temperature as compared to MLR and if compared different techniques of ANN then Multiple layer Artificial Neural Network (MLANN) proved to be best as compared to Single layer Artificial Neural Network (SLANN).