A Hybrid Photovoltaic Power Prediction Model Based on Multi-source Data Fusion and Deep Learning

With the rapid development of the world economy, environmental and energy problems have become increasingly prominent. This fact motivates the exploitation of renewable energy, such as solar power. Solar photovoltaic (PV) power has been integrated into modern smart grids on a large scale. Accurate photovoltaic prediction is the key to ensure the stable operation of the power grid. The single prediction model is difficult to be universal due to the limitations of the model. Therefore, this paper proposes a hybrid ultra-short-term photovoltaic power prediction model based on multi-source data fusion and deep learning. First, the satellite cloud images, historical power sequences, and numerical weather prediction (NWP) data are fused. Then, the photovoltaic power prediction models are established by using appropriate deep learning methods (convolutional neural networks (CNN), long-short term memory (LSTM) networks, and Extreme Gradient Boosting (XGBoost)) according to different data. Finally, the three models are combined to get the final prediction result. The results show that the proposed method in this paper has better forecasting performance than other single models.