A Review and Evaluation of Current Wind Power Prediction Technologies

The wind power prediction plays an essential role in operation, planning, taking part in open access and real time balancing of power system. Various forecasting methodologies have been proposed in number of research papers since last few decades. Therefore, on the basis of available literature, this review analyses new and current developments in the area of wind power & prediction of its derivatives (speed or direction) and compared in the form of comparative tables concerning the accuracy with taken care of variables to be predicted, time horizon, specific application area, data pre-processing, input data selection techniques, data used and various neural network techniques with their structure. The main focus of this review is to facilitate the various issues related to wind power forecasting techniques, emphasis on reduction of complexity of forecasting problem with increase in forecasting accuracy for different time span. The purpose of this research article is to motivate the power system researchers for designing new highly accurate online/offline models with concern to different issues regarding wind power resulting in secure reliable power system operation & better utilisation of energy resources. It has been observed that from a comparative forecasting accuracy analysis, hybrid models presented more accurate results as compared to other models. Key-Words: Wind power forecasting, time series, artificial intelligent techniques, accuracy criteria