Photovoltaic Power Prediction Algorithm Based on Wavelet Analysis and Neural Networks

Abstract

PV(Photovoltaic)power prediction could optimize generation planning and backup capacity configuration of the system, reduce the costs of power system operation and improve the economy and security of the operation. With the large-scale PV power plants accessing to power systems, the volatility of output power affects the operation and control of the power system, which is to be concerned more, and the academia and engineering focus on how to predict the output power more accurately and take effective measures. Since the periodic of the output power shows the non-stationary characteristics, this paper analyzes the factors which affects the output power, through a method which is a combination of wavelet analysis and neural network algorithm to improve the traditional way of forecasting, and improve the prediction accuracy of power prediction and then plays a great role in practical applications.