Photovoltaic Power Prediction Based on EMD-BLS Model

Abstract While the energy consumption structure is changing in this world, rate of renewable gradually increasing. This paper takes photovoltaic power as research object, analyses curve characteristics different climatic change situation and correlation between weather factors photovoltaic(PV) output, then puts forward a prediction model based on empirical mode decomposition-broad learning system (EMD-BLS). First, PV historical sequence needs to be reconstructed after pretreatment, EMD method used decompose output sequence. The broad established for each subsequence. Finally, results sub-series are superimposed get generation. tested with genuine data generation region our country. Compared traditional (BLS), extreme machine (ELM), LSTM, other models, error proposed lower, which can effectively improve accuracy