Photovoltaic power prediction model based on empirical mode decomposition-long-short memory neural network

Abstract

Listen

Translate article

The permeability of photovoltaic power supply in the power grid is constantly improving. Accurate short-term photovoltaic power generation prediction is conducive to ensuring the safe and stable operation of the power grid connected with high-ratio photovoltaic power supply. It is necessary to further improve the prediction accuracy of photovoltaic power in the process of power development. Photovoltaic output power will be affected by weather conditions and seasons, and it is easy to change with external factors, showing volatility. Therefore, with the introduction of a variety of influencing factors and increasing reference data, accurate and efficient short-term photovoltaic power generation prediction is conducive to ensuring the safe and stable operation of the power grid connected with high-ratio photovoltaic power supply. In order to further improve the prediction accuracy of photovoltaic power generation, this paper adopts a short-term rolling prediction method of photovoltaic power based on empirical mode decomposition and deep learning. A photovoltaic power prediction model combining empirical mode decomposition with long-term and short-term memory networks is proposed. The error comparison between BP, SVM and LSTM models and EMDLSTM model shows that EMD-LSTM has the highest prediction accuracy. The results can further improve the accuracy of photovoltaic power prediction.