Probabilistic Prediction of Wind Power Based on Gaussian Process Regression

Wind power prediction is commonly deterministic point prediction at present, due to the actual operating environment restriction and other artificial factors, such as hardware condition prediction accuracy increase the space is lesser, a scientific guidance on the grid and the operation of the wind farm, to maximize the use of wind power, reduce wind power fluctuation influence to power grid, this paper proposed wind power prediction method based on gaussian process regression, Combined with the change trend of measured wind power data obtained from actual wind farm plants, the appropriate covariance function is selected to obtain the probabilistic prediction result containing uncertain information, and the prediction accuracy of wind power is as high as 93.33%, which has good practical significance and practical value.