Grey predictor models for wind speed-wind power prediction

Abstract

High variability in generated wind power that results from the wind speed is the key factor that impacts power quality .B ased on grey predictor models, t his paper presents one-step to four-step average ten-minute wind speed forecasting and gives the residual error for steady wind. Wind speed predictions for unsteady wind and gust are also made. T hen, taking the steady wind speed for instance , fitting parameters in various models that wind power changes with wind speed are obtained by modelling the relationship between real wind power and sequential wind speed. In order to enhance the forecasting precision of wind power, the function model fitting for wind power characteristic is established through comparing the precision of different models from the aspect of piecewise function and overall model .T his paper compares the two modelling methods and finds the appropriate one .Compared the simulated value with actual value, better model fitting for wind power characteristics is obtained. F inally ,the actual data from wind farm in China proves that this method is effective. This work is supported by Special Fund of the National Basic Research Program of China (No.2009CB219702).