A Novel Efficient DLUBE Model Constructed by Error Interval Coefficients for Clustered Wind Power Prediction

Interval prediction is essential to improve the scheduling and planning of wind power systems. In this study, a novel lower upper bound estimation model based on gated recurrent unit was proposed for clustered forecasting. Different from existing research, directly realizes interval point results corresponding error coefficients, an unsupervised learning strategy introduced construct coefficients. addition, loss functions related characteristics are designed, effective gradient descent algorithm adopted optimize entire model. comparative experiments, two data were collected as experimental data, seven representative models selected benchmark models, which fully proved superiority