Power Generation Prediction Model of Distributed Photovoltaic System Based on Adaptive Extended Kalman Filter

Low-polluting renewable energy generation has expanded both in height and breadth over the past few years. Photovoltaic power generation is another widely used renewable energy generation technology after wind power generation. How-ever, due to the influence of multiple external factors, photovoltaic power generation has the disadvantages of uncertainty and intermittent of power generation, which increase the difficulty of grid connection and lead to the serious phenomenon of power abandonment. In order to solve these problems, this paper proposes a photovoltaic power generation prediction model to provide a reference for power distribution, grid connection decisions, and photovoltaic builders. Firstly, the solar insolation, which mainly affects the distributed photovoltaic generation is normalized, and the finite element calculation method is used to divide the solar radiation intensity in the study area, and the average radiation intensity in the cell is obtained to eliminate the influence of irrelevant environmental factors. Secondly, a new photovoltaic array arrangement method is proposed that has the advantages of a small footprint, long maximum radiation time, and the highest power generation. Finally, in order to obtain an accurate PV power generation prediction value, this paper devotes the adaptive extended Kalman filter algorithm to calculate the future PV power generation of the local place according to the existing power generation. The experimental results show that the prediction accuracy of the new filtering algorithm is higher than that of the prediction method based on the traditional extended Kalman filtering algorithm.