Multi-scale Forecasting Of Wind Speed And Output Power On The Wind Farm

Wind energy as a non-polluting and renewable energy, has attracted great attention over the world and become one of the rapid developed form of power generation. The main method of exploitation and utilization of wind energy is the large-scale power generation connected with grid at present. Wind speed and power prediction is an essential issue for wind farm maintenance, optimal power flow between power network and wind farms, electricity marketing bidding, power system scheduling, and energy storage planning and scheduling.Wind power as an intermittent energy has strong randomness and is difficult to be controlled, and usually, the fluctuation range of speed and the output power is very large. Because of these reasons, it is difficult to control the voltage, reactive power and peak power, presenting challenges for security, stability and scheduling of the power grid. When the wind power installed capacity reaches a certain proportion, the effective way to solve this problem is the accurate prediction of the output power in wind farms. Wind power forecasting can be divided by the time scales, which include short-term forecasting, medium-term forecasting and long-term forecasting. However, the presupposition of the accurate forecast of the wind power is accurate forecast of the wind speed. Considering the characteristics of wind speed, a short term wind speed forecasting method based on empirical mode decomposition (EMD) and least square support vector machine (LSSVM) method has been proposed; a medium-term wind speed forecasting model is based on wavelet decomposition (WT) and LSSVM method has been proposed; a long-term wind speed forecasting model is based on empirical mode decomposition (EMD) and recursive least square (RLS) method has been proposed in this paper.According to the multi-scale wind speed forecasting values, considering the influencing factors and the losses of the wind farm, the short-term, medium-term and long-term output power forecast of the wind farm can be obtained.