Along with the energy shortage, and the severe energy supply security situation, new energy such as wind power is developing rapidly, and the proportion of wind power in the grid is continuously increasing. However, compared with the hydropower and thermal power, the wind power is intermittentărandomness and uncontrollable, large-scale wind power integrated into power system will bring severe challenges of power system safety operation and power quality. Wind power forecasting technology can guide the dispatching of grid and the production planning of wind farm effectively. Therefore, it is urgent to carry out in-depth study for wind power output prediction technology, and for wind power of comparatively accurate prediction.This paper studies short-term wind power prediction. The statistical rules of parameters from a 200 MW wind farm are acquired after pretreatment, and the paper analyses wind speedâs chaotic characteristics. First of all, Analyzed of wind turbine power output curve, as well as wind speed, wind direction and other factors on wind power. Secondly, use the direct prediction method, using wind speed, wind direction, temperature, barometric pressure and other meteorological data as input predictive model for short-term wind farm power is predicted. Established BP neural network forecasting model, RBF neural network forecasting model and RBF-BP portfolio neural network forecasting model for short-term wind power prediction, then compared prediction results. The simulation results proved that RBF-BP neural network model has higher prediction precision, has the ability to adapt to varying characteristics of the time, has the very good nonlinear mapping ability, can be used in wind power prediction and other similar prediction. |