Photovoltaic Power Prediction Method Based on Chaos Analysis and Long Short-Term Memory Neural Network
Targeting at the randomness and uncertainty in photovoltaic power series,a prediction method based on chaos analysis and long short-term memory neural network(LSTM)is proposed to improve the prediction accuracy of photovoltaic power in complex environment.Chaos analysis methods include ensemble empirical mode decomposition(EEMD)and phase space reconstruction.EEMD is used to decom-pose photovoltaic power sequence and reduce the complexity of the original sequence.Phase space reconstruction reconstructs the original one-dimensional sequence into a sequence in high-dimensional state space,which is conducive to improving the prediction accuracy.LSTM has a more complex mechanism,which is more suitable for predicting uncertain time series.In addition,photovoltaic power is divided into stationary and non-stationary data,and prediction models for these two kinds of data are established.The experimental results show that the proposed method has different advantages over the single type prediction model and other combined models.
photovoltaic powerchaos analysisLSTMEEMDphase space reconstruction
万方数据
维普
