Multi-step Prediction of Ultra-short-term Wind Power Based on Non-stationary Transformer
Targeting the problem of low accuracy in the multi-step prediction of wind power caused by the volatility and randomness in wind power prediction,the paper proposes a multi-step prediction model of ultra-short-term wind power based on non-stationary transformer.The Pearson correlation coefficient(PCC)and principal component analysis(PCA)are used to analyze the wind power and its influencing factors to determine the input data.Based on the non-stationary transformer model that can enhance the effect of non-stationary time series prediction,the complex relationship between the input data and the output power is efficiently and adequately explored,and the ultra-short-term prediction model of the wind power is constructed.The example analysis shows that the proposed method has high prediction accuracy and more stable prediction results in predicting the wind power with different prediction step lengths.
wind powerpredictionPCCPCAnon-stationary transformer model
NETL
NSTL
万方数据
维普
