One Parametric Approach for Short-Term JPDF Forecast of Wind Generation

The time-domain correlation information of wind generation is important for wind power utilization. This paper proposes a parametric approach for the short-term multiperiod joint probability density function (JPDF) forecast of wind generation. The approach makes a spot forecast of wind generation by using support vector machine (SVM), and the probability distribution of the SVM forecast error is estimated by sparse Bayesian learning (SBL) which assumes that the forecast error follows a Gaussian distribution. Then, the SVM forecast result is corrected. The correlation matrix of the forecast errors within multiple successive forecast periods is estimated using the historical data. By combining the variance information obtained by SBL and the correlation matrix, the covariance matrix of the forecast errors is formed. Thereby, the JPDF of wind generation is obtained. Data collected from an actual wind farm are used for the study, and the results illustrate the effectiveness of the proposed approach. The justification of the Gaussian assumption of SBL is also explained in this paper.