Parametrical needs for wind speed prediction using ANN

In last few decades there has been a upward trend towards sustainable energy generation and consumption. Effect is a direct result of depleting fossil fuel reserves as well as increasing global concerns regarding the ill-effects of ongoing power generation techniques. In this sector wind power generation tends to play a vital role. However, the arbitrary nature of wind resource is a major hindrance in its application. Wind speed forecasting enables us to know the future wind reserve at any target site, as well as helps us to handle power issues that could result as an inadvertent effect of wind power integration into the system. In this paper focus was on finding out the most influential parameters for wind speed prediction purposes. These parameters when incorporated into any Numerical Weather Prediction (NWP) model would lead to best prediction results. Results have been verified over a time horizon of 6 and 12 hours ahead using 1 and 2 year models, respectively. Data from VABB airport, Mumbai was used for modelling and prediction purposes.