Exploration and Practice of a New Elastic Scheduling Model Based on Distributed Photovoltaic Grid Load Prediction

This paper proposes a distributed photovoltaic output prediction method based on spatial correlation and improved weather label. The distributed photovoltaic lines were sorted out, and the spatial correlation groups with the centralized photovoltaic stations were established. Combined with the weather of the forecast day and the improved weather mapping, the output curve under the multi-weather label was predicted by relying on the forecast model of the centralized photovoltaic stations. The optimal curve was selected by the Euclidean-distance algorithm, and finally the distributed photovoltaic output to be predicted was obtained by empirical formula. The output prediction model with improved weather label was adopted to calculate the distributed photovoltaic fine-tuning node capacity and optimize the maintenance plan under special circumstances, so as to explore and practice a new elastic scheduling model based on distributed photovoltaic grid load prediction.