Output prediction of wind power generation system using complex-valued neural network

In this paper, we proposed an output prediction system of a wind power generation using a complex-valued neural network (called here, CVNN). The CVNN is suitable for treating complex numbers and nonlinear data. Then we express the wind information (wind speed and direction) by complex numbers on the complex coordinates, and use them as input information for the CVNN. As input wind speed data, we use the numerical weather prediction data which are provided by Japan Meteorological Agency. Several simulation results demonstrate the effectiveness of the proposed prediction system.