Short-term solar photovoltaic irradiation predicting using a nonlinear prediction method

The power output of grid-connected solar photovoltaic (PV) is intermittent due to the influences of solar irradiance which makes the PV system unreliable especially in the condition of high penetration rates of PV. Forecasting the solar irradiance and power output of a PV plant is an effective way to solve the problem. The proposed method utilizes a tracking camera to capture the stochastic clouds near the sun. The predicting of PV power is achieved in two stages: in the first stage a series of image processing methods are used for calculating the motion of clouds and in the second stage a nonlinear predicting method is used for predicting the possible path of clouds and arrival time of clouds shading the PV system. The results show the effectiveness of the integrated method.