An Improved Method for Photovoltaic Forecasting Model Training Based on Similarity

Photovoltaic (PV) power generation is the most widely adopted renewable energy source. However, its inherent unpredictability poses considerable challenges to management of grids. To address arduous and time-consuming training process PV prediction models, which has been a major focus prior research, an improved approach for based on neighboring days proposed in this study. This specifically designed handle preprocessing datasets by leveraging results similarity analysis generation. Experimental demonstrate that method can significantly reduce time models without sacrificing accuracy, be effectively applied both ensemble deep learning approaches.