An ultra-short-term forecast method for photovoltaic power generation based on ResNet

Abstract Predicting photovoltaic power generation is important for enhancing the operation and management of photovoltaic power systems, as well as boosting their electricity production efficiency. An ultra-short-term photovoltaic power generation prediction model based on ResNet is proposed to address the current issues of unsatisfactory ground-based cloud image prediction performance and unfavorable marginalization deployment. After analyzing and processing the sky image, the ResNet network is used for feature extraction. Historical photovoltaic power generation data is used as input and fused with the results of feature extraction. Finally, the photovoltaic power generation power is predicted. The result of the experiment showed that the model proposed can effectively extract feature information for predicting ultra-short-term photovoltaic power generation.