Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
This study introduces a step-adaptive multimodal fusion network designed to improve ultra-short-term solar irradiance forecasting by addressing the limitations of existing models. By integrating InceptionNeXt for multi-scale cloud feature extraction and a dynamic low-frequency compensation unit, the model effectively captures both spatial and temporal dependencies in cloud dynamics. Experimental results demonstrate significant improvements in prediction accuracy over state-of-the-art methods, highlighting the model's practical applicability in photovoltaic system management.
A novel approach to solar irradiance forecasting achieves unprecedented accuracy by dynamically adapting to cloud conditions and temporal changes.
Ultra-short-term solar irradiance prediction is critical for photovoltaic system dispatch and power grid stability. Existing approaches suffer from three key shortcomings: single time-series models cannot capture the spatial dynamics of clouds under complex conditions, standard convolutions inadequately represent multi-scale cloud features, and fixed low-frequency compensation strategies fail to adapt to different prediction steps. To address these issues, this proposes a multi-source data fusion model for ultra-short-term irradiance prediction. The model first employs InceptionNeXt to extract multi-scale, multi-directional spatial features from ground-based cloud images. A step-adaptive low-frequency compensation unit is then introduced to dynamically modulate global low-frequency information based on the prediction step. Eventually, the enhanced image features are combined with meteorological time-series features, and a TempAttnLSTM network captures global temporal dependencies for multi-step prediction. Experiments on the public NREL dataset and practical photovoltaic stations in Shandong illustrate the effectiveness of the proposed method compared with several state-of-the-art approaches.