Wind energy is a renewable energy sources where numerous researches on wind energy forecasting have been carried out to construct efficient forecasting models. Statistical approaches like Auto Regressive Integrated Moving Average (ARIMA), Exponential smoothing are used to handle short-term wind speed forecasting. Artificial Neural Networks (ANN) based algorithms like Multilayer Perceptron (MLP), Recurrent Neural Network (RNN) algorithms are used to capture the nonlinear relationship between the input and output information. We have chosen the dataset acquired from a wind turbine at TURKEY with 10-minute time interval and forecasting performance are verified with actual data. This work considers wind speed as the input parameter and built univariate model using statistical techniques such as ARIMA, Exponential smoothing etc. To improve the forecasting accuracy, a multivariate model (wind speed and direction) is built using various intelligent learning based algorithms (ANN). ANN algorithms such as random forest algorithm, LSTM algorithm (Univariate, Multivariate) are implemented on wind turbine time series data to conduct an experimental study on statistical and artificial neural network based algorithms. Metrics like MSE, RMSE are considered and experiment results shows that ANN based algorithms provides accurate forecasting results.