Short-term wind power forecasting by genetic algorithm of wavelet neural network

Abstract

Wavelet neural network (WNN) is widely used in wind power prediction because of its good self-learning capacity and excellent performance to approach any nonlinear function. However it also has limitation in precision and operating speed. This paper proposes a new genetic algorithm of wavelet neural network (GAWNN) for short-term wind power forecasting in electrical power systems. GAWNN makes a good combination of genetic algorithm and wavelet neural network and makes great process in convergent precision and speed. The experiment results show that GAWNN is more feasible and effective.