Short-term Wind Power Forecasting Based on Convolutional Neural Networks

Jinli Dou,Chun Liu,Bo Wang
DOI: https://doi.org/10.1088/1755-1315/170/4/042023
2018-07-01
IOP Conference Series: Earth and Environmental Science
Abstract:Traditional methods of short-term wind power prediction are mostly based on NWP (Numerical Weather Prediction) data on a single station in single-time cross-section and lack in spatiotemporal correlation mining of data. Therefore, a CNN-based prediction method is proposed. Firstly, based on the theoretical analysis of convolution neural network, the input was modelled considering the time correlation, and a variety of convolution neural network structures were designed. Then, a variety of error evaluation criteria were used to evaluate the correlation between single-layer and multilayer feedforward neural networks as well as the convolutional neural networks prediction method. The error and the actual prediction results were analyzed. Prediction error analysis shows that the convolution neural network model can effectively mine the time correlation between data and improve the accuracy of short-term wind power predictions.
What problem does this paper attempt to address?