Wind power forecasting is the basic and essential task in the wind power plant operation. Wind power is one of the fast-growing feasible power resources and it will be viewed as the additional substitute for conservative power produced from the non-sustainable power source. The use of the conservative source of electricity will be reduced with the help of wind power forecasting (WPF). The fluctuation and instability of the wind will be difficult for collecting the original data and it will the major impact for forecasting the accuracy. Last few years many researchers use different data mining technique that has been applied in various prediction system that produced good forecasting accuracy. To improve the accuracy of the forecasting and reduce the computational complexity for a hybrid method is proposed that consist of fuzzy k-means clustering and Bagging Neural Network (BNN). In historical days there is a lot of similarities, fuzzy k-means clustering is used for clustering the similar days it consisting of the detailed information about historical data and weather conditions. In order to avoid the overfitting problem, a bagging algorithm to be incorporated with backpropagation neural network. For proving the efficiency, the proposed hybrid method will be evaluated in real wind farm which will provide better forecasting accuracy and also expected to reduce the computational complexity when compared with other existing wind power forecasting approaches.