Short-term photovoltaic power prediction method and system based on hybrid model

The invention provides a short-term photovoltaic power prediction method and system based on a hybrid model. The method comprises the following steps: acquiring data related to photovoltaic power; constructing a hybrid model of a three-dimensional convolution and convolution long-short-term memory neural network; training the hybrid model; and obtaining a short-term photovoltaic power prediction result based on the data related to the photovoltaic power by using the trained hybrid model. The prediction result of the method is higher in precision, the consumed time is short, and the applicability, accuracy and stability of the established model to photovoltaic power prediction are proved. 本发明提供了一种基于混合模型的短期光伏功率预测方法及系统，获取和光伏功率相关的数据；构建三维卷积和卷积长短期记忆神经网络的混合模型；训练所述混合模型；利用训练后的混合模型，基于和光伏功率相关的数据，得到短期光伏功率预测结果。本发明的预测结果不仅精度更高，且耗时较短，证实了所建模型对光伏功率预测的适用性、准确性、和稳定性。