Short-term wind power forecasting using a double-stage hierarchical hybrid GA-ANFIS approach

Wind generation power output estimation is always associated with some uncertainties as a result of wind speed and other weather parameters intermittency, and accurate short-term predictions are important for their efficient operation. This can greatly help transmission and distribution system operators and schedulers to improve the power network control and management. In this paper, a double stage hierarchical genetic algorithm based adaptive neuro-fuzzy inference system (double-stage hybrid GA-ANFIS) approach is proposed for short-term wind power forecast of a microgrid wind farm in Beijing, China. The approach has two hierarchical stages. The first GA-ANFIS stage utilizes numerical weather prediction (NWP) meteorological parameters to predict wind speed at the wind farm exact site and turbine hub height. The second stage maps the actual wind speed and power relationships. Then, the forecasted next day's wind speed by the first stage is applied to the second stage to predict next day's wind power. The presented approach has achieved considerable prediction accuracy enhancement. The accuracy of the proposed model is compared with other four forecasting methods and resulted in the best accuracy improvement of all.