Photovoltaic power prediction based on similar day clustering and Transformer-BiLSTM

Accurate prediction of photovoltaic power has become a core research topic in the field of power systems. To improve the prediction accuracy of photovoltaic power generation, this paper proposes a photovoltaic power prediction method based on similar day clustering and Transformer-BiLSTM. Firstly, this paper normalizes the data, selects strongly correlated meteorological parameter features as the model input through the Spearman correlation coefficient, selects the optimal number of clusters based on the BIC criterion and selects similar days through K-means, and then inputs the data into the Transformer-BiLSTM prediction model. Finally, the measured data of a certain photovoltaic power plant is studied and compared with a single prediction model. The results show that the proposed model has higher prediction accuracy.