Short term wind forecasting using logistic regression driven hypothesis in artificial neural network

The share of wind power is increasing significantly all over the world. The ever increasing wind power integration poses new issues due to its variability and volatility. Good forecasting techniques are thus important to address these challenges. In this paper, few time series forecasting models like artificial neural networks, adaptive neuro fuzzy interface systems are used for short term prediction of wind speeds and further a new hypothesis for better estimation of wind speed is proposed. The results obtained from a real world case study of a wind farm in the state of Karnataka are presented. In this experimental study, a thorough investigation is carried out, considering the results obtained from the mentioned techniques, the accuracy of the proposed model is found to be better by 13.53% than the existing techniques.